davanstrien/model_cards_with_readmes · Datasets at Hugging Face

Helsinki-NLP/opus-mt-it-ar

Helsinki-NLP

marian

11

35

transformers

0

translation

true

false

apache-2.0

['it', 'ar']

null

1

0

['translation']

false

true

2,034

false

### ita-ara * source group: Italian * target group: Arabic * OPUS readme: [ita-ara](https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/ita-ara/README.md) * model: transformer * source language(s): ita * target language(s): ara * model: transformer * pre-processing: normalization + SentencePiece (spm32k,spm32k) * download original weights: [opus-2020-07-03.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/ita-ara/opus-2020-07-03.zip) * test set translations: [opus-2020-07-03.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/ita-ara/opus-2020-07-03.test.txt) * test set scores: [opus-2020-07-03.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/ita-ara/opus-2020-07-03.eval.txt) ## Benchmarks | testset | BLEU | chr-F | |-----------------------|-------|-------| | Tatoeba-test.ita.ara | 21.9 | 0.517 | ### System Info: - hf_name: ita-ara - source_languages: ita - target_languages: ara - opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/ita-ara/README.md - original_repo: Tatoeba-Challenge - tags: ['translation'] - languages: ['it', 'ar'] - src_constituents: {'ita'} - tgt_constituents: {'apc', 'ara', 'arq_Latn', 'arq', 'afb', 'ara_Latn', 'apc_Latn', 'arz'} - src_multilingual: False - tgt_multilingual: False - prepro: normalization + SentencePiece (spm32k,spm32k) - url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/ita-ara/opus-2020-07-03.zip - url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/ita-ara/opus-2020-07-03.test.txt - src_alpha3: ita - tgt_alpha3: ara - short_pair: it-ar - chrF2_score: 0.517 - bleu: 21.9 - brevity_penalty: 0.95 - ref_len: 1161.0 - src_name: Italian - tgt_name: Arabic - train_date: 2020-07-03 - src_alpha2: it - tgt_alpha2: ar - prefer_old: False - long_pair: ita-ara - helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 - transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b - port_machine: brutasse - port_time: 2020-08-21-14:41

a955cc4eae22416ff8d45b324a426d43

cartesinus/xlm-r-base-amazon-massive-intent-label_smoothing

cartesinus

xlm-roberta

11

6

transformers

0

text-classification

true

false

mit

['en']

['AmazonScience/massive']

null

0

['generated_from_trainer', 'nlu', 'intent-classification', 'text-classification']

true

1,639

false

# xlm-r-base-amazon-massive-intent-label_smoothing This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the [MASSIVE1.1](https://huggingface.co/datasets/AmazonScience/massive) dataset. It achieves the following results on the evaluation set: - Loss: 2.5148 - Accuracy: 0.8879 - F1: 0.8879 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 - label_smoothing_factor: 0.4 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:| | 3.3945 | 1.0 | 720 | 2.7175 | 0.7900 | 0.7900 | | 2.7629 | 2.0 | 1440 | 2.5660 | 0.8549 | 0.8549 | | 2.5143 | 3.0 | 2160 | 2.5389 | 0.8711 | 0.8711 | | 2.4678 | 4.0 | 2880 | 2.5172 | 0.8883 | 0.8883 | | 2.4187 | 5.0 | 3600 | 2.5148 | 0.8879 | 0.8879 | ### Framework versions - Transformers 4.24.0 - Pytorch 1.12.1+cu113 - Datasets 2.7.0 - Tokenizers 0.13.2

6d959a603f0dec9ef7306cd9341f6edd

jerpint/whisper

jerpint

null

5

0

null

2

translation

false

mit

['en']

['whisper']

null

0

['translation', 'speech', 'audio', 'automatic-speech-recognition']

false

true

2,489

false

This model was forked from the original [OpenAI whisper model](https://github.com/openai/whisper). # Whisper ## Model Whisper is a multi-lingual speech-to-text model. It takes in raw audio recordings from many languages and outputs transcriptions in the language of origin or translated to english. The model first converts speech to spectrograms, then uses an auto-regressive transformer to decode the speech to text. Here is an overview of the architecture: ![model_architecure](https://github.com/jerpint/whisper/raw/main/approach.png) For more information on the technical implementations, consult the [paper](https://cdn.openai.com/papers/whisper.pdf). ## Training Data The model was trained on 680 000 hours of audio and associated transcripts trained from the internet. The majority of the audio is in english (~65%) while the remainder is in other languages. A total of 98 different languages were used in the dataset. ![image](https://user-images.githubusercontent.com/18450628/204110014-e2684385-d790-4dd7-8ce1-47168efb2726.png) ## Model Variations OpenAI has released 9 different versions of the model, trained either on english-only audio or on multilingual data. | Size | Parameters | English-only model | Multilingual model | Required VRAM | Relative speed | |:------:|:----------:|:------------------:|:------------------:|:-------------:|:--------------:| | tiny | 39 M | `tiny.en` | `tiny` | ~1 GB | ~32x | | base | 74 M | `base.en` | `base` | ~1 GB | ~16x | | small | 244 M | `small.en` | `small` | ~2 GB | ~6x | | medium | 769 M | `medium.en` | `medium` | ~5 GB | ~2x | | large | 1550 M | N/A | `large` | ~10 GB | 1x | ## Limitations and bias In the [paper](https://cdn.openai.com/papers/whisper.pdf), they find a direct corelation between performance on a given language and the amount of data available in the dataset. As such, languages that are under-represented in the scraped dataset perform less well in whisper. Because english is much more prevalent than other languages, the model will likely perform better in english. This is shown in the following figure, where a lower word error rate (WER) indicates a better performance: ![model_performance](https://github.com/jerpint/whisper/raw/main/language-breakdown.svg)

d066e4db24448b306d99575a34e97e7f

Anery/bert-finetuned-ner

Anery

bert

17

3

transformers

0

token-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,602

false

# bert-finetuned-ner This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.0244 - Precision: 0.7368 - Recall: 0.4 - F1: 0.5185 - Accuracy: 0.9919 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 4 ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| | No log | 1.0 | 14 | 0.0598 | 0.0 | 0.0 | 0.0 | 0.9870 | | No log | 2.0 | 28 | 0.0357 | 0.0 | 0.0 | 0.0 | 0.9894 | | No log | 3.0 | 42 | 0.0256 | 0.75 | 0.2571 | 0.3830 | 0.9910 | | No log | 4.0 | 56 | 0.0244 | 0.7368 | 0.4 | 0.5185 | 0.9919 | ### Framework versions - Transformers 4.19.2 - Pytorch 1.11.0+cu113 - Datasets 2.2.2 - Tokenizers 0.12.1

f69413195abda97aa981bffe22bd3760

napatswift/bkk-ner-model

napatswift

bert

17

3

transformers

0

token-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

2,638

false

# bkk-ner-model This model is a fine-tuned version of [Geotrend/bert-base-th-cased](https://huggingface.co/Geotrend/bert-base-th-cased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.0518 - Precision: 0.8850 - Recall: 0.9615 - F1: 0.9217 - Accuracy: 0.9822 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 15 ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| | No log | 1.0 | 8 | 0.5592 | 0.3698 | 0.6827 | 0.4797 | 0.7818 | | No log | 2.0 | 16 | 0.4491 | 0.4831 | 0.8269 | 0.6099 | 0.8062 | | No log | 3.0 | 24 | 0.3738 | 0.6226 | 0.9519 | 0.7529 | 0.8399 | | No log | 4.0 | 32 | 0.1781 | 0.6691 | 0.8942 | 0.7654 | 0.9401 | | No log | 5.0 | 40 | 0.2201 | 0.8095 | 0.9808 | 0.8870 | 0.9204 | | No log | 6.0 | 48 | 0.0936 | 0.8130 | 0.9615 | 0.8811 | 0.9710 | | No log | 7.0 | 56 | 0.0692 | 0.8197 | 0.9615 | 0.8850 | 0.9757 | | No log | 8.0 | 64 | 0.0712 | 0.8264 | 0.9615 | 0.8889 | 0.9710 | | No log | 9.0 | 72 | 0.0575 | 0.8621 | 0.9615 | 0.9091 | 0.9803 | | No log | 10.0 | 80 | 0.0625 | 0.8487 | 0.9712 | 0.9058 | 0.9766 | | No log | 11.0 | 88 | 0.0580 | 0.8584 | 0.9327 | 0.8940 | 0.9766 | | No log | 12.0 | 96 | 0.0551 | 0.8684 | 0.9519 | 0.9083 | 0.9813 | | No log | 13.0 | 104 | 0.0554 | 0.8761 | 0.9519 | 0.9124 | 0.9803 | | No log | 14.0 | 112 | 0.0535 | 0.8772 | 0.9615 | 0.9174 | 0.9813 | | No log | 15.0 | 120 | 0.0518 | 0.8850 | 0.9615 | 0.9217 | 0.9822 | ### Framework versions - Transformers 4.25.1 - Pytorch 1.13.0+cu116 - Datasets 2.8.0 - Tokenizers 0.13.2

4af13549d921773bb4ed81df28a329ca

EvSz/PokemonDiffuser-128

EvSz

null

18

7

diffusers

1

null

false

apache-2.0

['en']

['EvSz/Pokemon-by-Name-512px']

null

0

[]

false

true

1,201

false

# PokemonDiffuser-128 ## Model description This diffusion model is trained with the [🤗 Diffusers](https://github.com/huggingface/diffusers) library on the `huggan/pokemon` dataset. ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training data [TODO: describe the data used to train the model] ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 16 - eval_batch_size: 16 - gradient_accumulation_steps: 1 - optimizer: AdamW with betas=(None, None), weight_decay=None and epsilon=None - lr_scheduler: None - lr_warmup_steps: 500 - ema_inv_gamma: None - ema_inv_gamma: None - ema_inv_gamma: None - mixed_precision: fp16 ### Training results 📈 [TensorBoard logs](https://huggingface.co/EvSz/PokemonDiffuser-128/tensorboard?#scalars)

3b0bd4df4537ea34234f529611239c2c

AymanMansour/Whisper-Sudanese-Dialect-lsrge-v2-10K

AymanMansour

whisper

41

22

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,854

false

# openai/whisper-large-v2 This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.0925 - Wer: 41.4086 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 8 - eval_batch_size: 4 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - training_steps: 10000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:-----:|:---------------:|:-------:| | 0.5216 | 1.04 | 1000 | 0.7054 | 58.7611 | | 0.0872 | 3.02 | 2000 | 0.7803 | 60.1400 | | 0.1073 | 4.06 | 3000 | 0.8312 | 61.0522 | | 0.0617 | 6.04 | 4000 | 0.8583 | 48.2181 | | 0.0053 | 8.02 | 5000 | 0.9135 | 41.8328 | | 0.0049 | 9.06 | 6000 | 0.9697 | 43.3814 | | 0.0044 | 11.04 | 7000 | 0.9863 | 41.9813 | | 0.0006 | 13.02 | 8000 | 1.0359 | 42.7662 | | 0.0019 | 14.06 | 9000 | 1.0714 | 41.3449 | | 0.0007 | 16.04 | 10000 | 1.0925 | 41.4086 | ### Framework versions - Transformers 4.26.0.dev0 - Pytorch 1.13.1+cu117 - Datasets 2.7.1.dev0 - Tokenizers 0.13.2

d2fc164a2763ecb4230c960539f419e2

JovialValley/model_broadclass_onSet0.1

JovialValley

wav2vec2

13

0

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

13,085

false

# model_broadclass_onSet0.1 This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1129 - 0 Precision: 1.0 - 0 Recall: 1.0 - 0 F1-score: 1.0 - 0 Support: 31 - 1 Precision: 0.9259 - 1 Recall: 1.0 - 1 F1-score: 0.9615 - 1 Support: 25 - 2 Precision: 1.0 - 2 Recall: 0.9259 - 2 F1-score: 0.9615 - 2 Support: 27 - 3 Precision: 1.0 - 3 Recall: 1.0 - 3 F1-score: 1.0 - 3 Support: 15 - Accuracy: 0.9796 - Macro avg Precision: 0.9815 - Macro avg Recall: 0.9815 - Macro avg F1-score: 0.9808 - Macro avg Support: 98 - Weighted avg Precision: 0.9811 - Weighted avg Recall: 0.9796 - Weighted avg F1-score: 0.9796 - Weighted avg Support: 98 - Wer: 0.0859 - Mtrix: [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 0, 25, 0, 0], [2, 0, 2, 25, 0], [3, 0, 0, 0, 15]] ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 16 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 200 - num_epochs: 80 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | 0 Precision | 0 Recall | 0 F1-score | 0 Support | 1 Precision | 1 Recall | 1 F1-score | 1 Support | 2 Precision | 2 Recall | 2 F1-score | 2 Support | 3 Precision | 3 Recall | 3 F1-score | 3 Support | Accuracy | Macro avg Precision | Macro avg Recall | Macro avg F1-score | Macro avg Support | Weighted avg Precision | Weighted avg Recall | Weighted avg F1-score | Weighted avg Support | Wer | Mtrix | |:-------------:|:-----:|:----:|:---------------:|:-----------:|:--------:|:----------:|:---------:|:-----------:|:--------:|:----------:|:---------:|:-----------:|:--------:|:----------:|:---------:|:-----------:|:--------:|:----------:|:---------:|:--------:|:-------------------:|:----------------:|:------------------:|:-----------------:|:----------------------:|:-------------------:|:---------------------:|:--------------------:|:------:|:---------------------------------------------------------------------------------------:| | 2.343 | 4.16 | 100 | 2.2083 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 2.2769 | 8.33 | 200 | 2.1649 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 1.9687 | 12.49 | 300 | 1.8723 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 1.8046 | 16.65 | 400 | 1.6982 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 1.5645 | 20.82 | 500 | 1.5862 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 1.5322 | 24.98 | 600 | 1.5736 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 1.5468 | 29.16 | 700 | 1.4736 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 1.0542 | 33.33 | 800 | 1.0068 | 0.3163 | 1.0 | 0.4806 | 31 | 0.0 | 0.0 | 0.0 | 25 | 0.0 | 0.0 | 0.0 | 27 | 0.0 | 0.0 | 0.0 | 15 | 0.3163 | 0.0791 | 0.25 | 0.1202 | 98 | 0.1001 | 0.3163 | 0.1520 | 98 | 0.9847 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 25, 0, 0, 0], [2, 27, 0, 0, 0], [3, 15, 0, 0, 0]] | | 0.9664 | 37.49 | 900 | 0.9831 | 0.3483 | 1.0 | 0.5167 | 31 | 1.0 | 0.12 | 0.2143 | 25 | 1.0 | 0.0370 | 0.0714 | 27 | 0.8 | 0.2667 | 0.4 | 15 | 0.3980 | 0.7871 | 0.3559 | 0.3006 | 98 | 0.7632 | 0.3980 | 0.2990 | 98 | 0.9758 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 21, 3, 0, 1], [2, 26, 0, 1, 0], [3, 11, 0, 0, 4]] | | 0.9405 | 41.65 | 1000 | 0.9402 | 0.3827 | 1.0 | 0.5536 | 31 | 1.0 | 0.04 | 0.0769 | 25 | 1.0 | 0.4815 | 0.65 | 27 | 1.0 | 0.2 | 0.3333 | 15 | 0.4898 | 0.8457 | 0.4304 | 0.4035 | 98 | 0.8047 | 0.4898 | 0.4248 | 98 | 0.9630 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 24, 1, 0, 0], [2, 14, 0, 13, 0], [3, 12, 0, 0, 3]] | | 0.9341 | 45.82 | 1100 | 0.9330 | 0.5082 | 1.0 | 0.6739 | 31 | 0.9231 | 0.48 | 0.6316 | 25 | 1.0 | 0.6296 | 0.7727 | 27 | 0.8571 | 0.4 | 0.5455 | 15 | 0.6735 | 0.8221 | 0.6274 | 0.6559 | 98 | 0.8029 | 0.6735 | 0.6707 | 98 | 0.9497 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 12, 12, 0, 1], [2, 9, 1, 17, 0], [3, 9, 0, 0, 6]] | | 0.8769 | 49.98 | 1200 | 0.8662 | 0.6327 | 1.0 | 0.775 | 31 | 0.9565 | 0.88 | 0.9167 | 25 | 1.0 | 0.6296 | 0.7727 | 27 | 0.8889 | 0.5333 | 0.6667 | 15 | 0.7959 | 0.8695 | 0.7607 | 0.7828 | 98 | 0.8557 | 0.7959 | 0.7939 | 98 | 0.9442 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 2, 22, 0, 1], [2, 9, 1, 17, 0], [3, 7, 0, 0, 8]] | | 0.8122 | 54.16 | 1300 | 0.7951 | 0.9062 | 0.9355 | 0.9206 | 31 | 0.8519 | 0.92 | 0.8846 | 25 | 1.0 | 0.8519 | 0.92 | 27 | 0.9375 | 1.0 | 0.9677 | 15 | 0.9184 | 0.9239 | 0.9268 | 0.9232 | 98 | 0.9230 | 0.9184 | 0.9185 | 98 | 0.9348 | [[0, 1, 2, 3], [0, 29, 2, 0, 0], [1, 1, 23, 0, 1], [2, 2, 2, 23, 0], [3, 0, 0, 0, 15]] | | 0.5747 | 58.33 | 1400 | 0.4843 | 1.0 | 1.0 | 1.0 | 31 | 0.96 | 0.96 | 0.96 | 25 | 1.0 | 0.9630 | 0.9811 | 27 | 0.9375 | 1.0 | 0.9677 | 15 | 0.9796 | 0.9744 | 0.9807 | 0.9772 | 98 | 0.9802 | 0.9796 | 0.9797 | 98 | 0.6732 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 0, 24, 0, 1], [2, 0, 1, 26, 0], [3, 0, 0, 0, 15]] | | 0.2794 | 62.49 | 1500 | 0.2062 | 1.0 | 1.0 | 1.0 | 31 | 0.96 | 0.96 | 0.96 | 25 | 1.0 | 0.9630 | 0.9811 | 27 | 0.9375 | 1.0 | 0.9677 | 15 | 0.9796 | 0.9744 | 0.9807 | 0.9772 | 98 | 0.9802 | 0.9796 | 0.9797 | 98 | 0.2236 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 0, 24, 0, 1], [2, 0, 1, 26, 0], [3, 0, 0, 0, 15]] | | 0.1654 | 66.65 | 1600 | 0.1573 | 1.0 | 0.9677 | 0.9836 | 31 | 0.9259 | 1.0 | 0.9615 | 25 | 1.0 | 0.9630 | 0.9811 | 27 | 1.0 | 1.0 | 1.0 | 15 | 0.9796 | 0.9815 | 0.9827 | 0.9816 | 98 | 0.9811 | 0.9796 | 0.9798 | 98 | 0.1303 | [[0, 1, 2, 3], [0, 30, 1, 0, 0], [1, 0, 25, 0, 0], [2, 0, 1, 26, 0], [3, 0, 0, 0, 15]] | | 0.1092 | 70.82 | 1700 | 0.1451 | 1.0 | 0.9677 | 0.9836 | 31 | 0.8889 | 0.96 | 0.9231 | 25 | 1.0 | 0.9259 | 0.9615 | 27 | 0.9375 | 1.0 | 0.9677 | 15 | 0.9592 | 0.9566 | 0.9634 | 0.9590 | 98 | 0.9621 | 0.9592 | 0.9597 | 98 | 0.1056 | [[0, 1, 2, 3], [0, 30, 1, 0, 0], [1, 0, 24, 0, 1], [2, 0, 2, 25, 0], [3, 0, 0, 0, 15]] | | 0.085 | 74.98 | 1800 | 0.1126 | 1.0 | 1.0 | 1.0 | 31 | 0.9259 | 1.0 | 0.9615 | 25 | 1.0 | 0.9259 | 0.9615 | 27 | 1.0 | 1.0 | 1.0 | 15 | 0.9796 | 0.9815 | 0.9815 | 0.9808 | 98 | 0.9811 | 0.9796 | 0.9796 | 98 | 0.0938 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 0, 25, 0, 0], [2, 0, 2, 25, 0], [3, 0, 0, 0, 15]] | | 0.0824 | 79.16 | 1900 | 0.1118 | 1.0 | 1.0 | 1.0 | 31 | 0.9259 | 1.0 | 0.9615 | 25 | 1.0 | 0.9259 | 0.9615 | 27 | 1.0 | 1.0 | 1.0 | 15 | 0.9796 | 0.9815 | 0.9815 | 0.9808 | 98 | 0.9811 | 0.9796 | 0.9796 | 98 | 0.0859 | [[0, 1, 2, 3], [0, 31, 0, 0, 0], [1, 0, 25, 0, 0], [2, 0, 2, 25, 0], [3, 0, 0, 0, 15]] | ### Framework versions - Transformers 4.25.1 - Pytorch 1.13.0+cu116 - Datasets 2.8.0 - Tokenizers 0.13.2

4274e11248901d38699e72590eaf2680

smeoni/nbme-electra-large-generator

smeoni

electra

17

2

transformers

0

text-generation

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,418

false

# nbme-electra-large-generator This model is a fine-tuned version of [google/electra-large-generator](https://huggingface.co/google/electra-large-generator) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.0122 - Accuracy: 0.9977 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3.0 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | No log | 1.0 | 195 | 0.1125 | 0.9789 | | No log | 2.0 | 390 | 0.0141 | 0.9973 | | 0.6233 | 3.0 | 585 | 0.0122 | 0.9977 | ### Framework versions - Transformers 4.19.0.dev0 - Pytorch 1.11.0 - Datasets 2.1.0 - Tokenizers 0.12.1

4f9bb7c233f4d8576a39d3000bb758fb

thkkvui/xlm-roberta-base-finetuned-panx-de-fr

thkkvui

xlm-roberta

10

3

transformers

0

token-classification

true

false

mit

null

0

['generated_from_trainer']

true

1,326

false

# xlm-roberta-base-finetuned-panx-de-fr This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1623 - F1: 0.8602 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 24 - eval_batch_size: 24 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:----:|:---------------:|:------:| | 0.2927 | 1.0 | 715 | 0.1798 | 0.8356 | | 0.1482 | 2.0 | 1430 | 0.1573 | 0.8507 | | 0.095 | 3.0 | 2145 | 0.1623 | 0.8602 | ### Framework versions - Transformers 4.20.1 - Pytorch 1.13.0.dev20220711 - Datasets 2.4.0 - Tokenizers 0.12.1

e146ec588e2af0ef5b02f9ab36fa4b62

theojolliffe/bart-large-cnn-finetuned-roundup-3-4

theojolliffe

bart

13

4

transformers

0

text2text-generation

true

false

mit

null

0

['generated_from_trainer']

true

1,777

false

# bart-large-cnn-finetuned-roundup-3-4 This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 1.1949 - Rouge1: 49.6216 - Rouge2: 29.1874 - Rougel: 32.042 - Rougelsum: 46.3679 - Gen Len: 140.9688 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 2 - eval_batch_size: 2 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 4 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:--------:| | No log | 1.0 | 258 | 1.2708 | 48.8914 | 29.2868 | 30.6203 | 46.2886 | 142.0 | | 1.1751 | 2.0 | 516 | 1.1869 | 49.3567 | 28.4751 | 31.3075 | 46.3408 | 141.75 | | 1.1751 | 3.0 | 774 | 1.1869 | 48.8335 | 28.4976 | 30.5434 | 46.2584 | 141.625 | | 0.7391 | 4.0 | 1032 | 1.1949 | 49.6216 | 29.1874 | 32.042 | 46.3679 | 140.9688 | ### Framework versions - Transformers 4.18.0 - Pytorch 1.11.0+cu113 - Datasets 2.1.0 - Tokenizers 0.12.1

aff5f7e820fdc63413912de93a712aa0

Helsinki-NLP/opus-mt-fi-sw

Helsinki-NLP

marian

10

8

transformers

0

translation

true

false

apache-2.0

null

0

['translation']

false

true

768

false

### opus-mt-fi-sw * source languages: fi * target languages: sw * OPUS readme: [fi-sw](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/fi-sw/README.md) * dataset: opus * model: transformer-align * pre-processing: normalization + SentencePiece * download original weights: [opus-2020-01-08.zip](https://object.pouta.csc.fi/OPUS-MT-models/fi-sw/opus-2020-01-08.zip) * test set translations: [opus-2020-01-08.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/fi-sw/opus-2020-01-08.test.txt) * test set scores: [opus-2020-01-08.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/fi-sw/opus-2020-01-08.eval.txt) ## Benchmarks | testset | BLEU | chr-F | |-----------------------|-------|-------| | JW300.fi.sw | 29.9 | 0.548 |

1926927dbd6245f4f3dd3f9a74182ec6

espnet/GunnarThor_talromur_a_fastspeech2

espnet

null

22

24

espnet

0

text-to-speech

false

cc-by-4.0

['en']

['talromur']

null

0

['espnet', 'audio', 'text-to-speech']

false

true

7,774

false

## ESPnet2 TTS model ### `espnet/GunnarThor_talromur_a_fastspeech2` This model was trained by Gunnar Thor using talromur recipe in [espnet](https://github.com/espnet/espnet/). ### Demo: How to use in ESPnet2 ```bash cd espnet git checkout 49a284e69308d81c142b89795de255b4ce290c54 pip install -e . cd egs2/talromur/tts1 ./run.sh --skip_data_prep false --skip_train true --download_model espnet/GunnarThor_talromur_a_fastspeech2 ``` ## TTS config <details><summary>expand</summary> ``` config: conf/tuning/train_fastspeech2.yaml print_config: false log_level: INFO dry_run: false iterator_type: sequence output_dir: exp/a/tts_train_fastspeech2_raw_phn_none ngpu: 1 seed: 0 num_workers: 1 num_att_plot: 3 dist_backend: nccl dist_init_method: env:// dist_world_size: null dist_rank: null local_rank: 0 dist_master_addr: null dist_master_port: null dist_launcher: null multiprocessing_distributed: false unused_parameters: false sharded_ddp: false cudnn_enabled: true cudnn_benchmark: false cudnn_deterministic: true collect_stats: false write_collected_feats: false max_epoch: 100 patience: null val_scheduler_criterion: - valid - loss early_stopping_criterion: - valid - loss - min best_model_criterion: - - valid - loss - min - - train - loss - min keep_nbest_models: 5 nbest_averaging_interval: 0 grad_clip: 1.0 grad_clip_type: 2.0 grad_noise: false accum_grad: 8 no_forward_run: false resume: true train_dtype: float32 use_amp: false log_interval: null use_matplotlib: true use_tensorboard: true use_wandb: false wandb_project: null wandb_id: null wandb_entity: null wandb_name: null wandb_model_log_interval: -1 detect_anomaly: false pretrain_path: null init_param: [] ignore_init_mismatch: false freeze_param: [] num_iters_per_epoch: 800 batch_size: 20 valid_batch_size: null batch_bins: 2500000 valid_batch_bins: null train_shape_file: - exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/train/text_shape.phn - exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/train/speech_shape valid_shape_file: - exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/valid/text_shape.phn - exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/valid/speech_shape batch_type: numel valid_batch_type: null fold_length: - 150 - 204800 sort_in_batch: descending sort_batch: descending multiple_iterator: false chunk_length: 500 chunk_shift_ratio: 0.5 num_cache_chunks: 1024 train_data_path_and_name_and_type: - - dump/raw/train_a_phn/text - text - text - - exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/train_a_phn/durations - durations - text_int - - dump/raw/train_a_phn/wav.scp - speech - sound valid_data_path_and_name_and_type: - - dump/raw/dev_a_phn/text - text - text - - exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/dev_a_phn/durations - durations - text_int - - dump/raw/dev_a_phn/wav.scp - speech - sound allow_variable_data_keys: false max_cache_size: 0.0 max_cache_fd: 32 valid_max_cache_size: null optim: adam optim_conf: lr: 1.0 scheduler: noamlr scheduler_conf: model_size: 384 warmup_steps: 4000 token_list: - <blank> - <unk> - ',' - . - r - t - n - a0 - s - I0 - D - l - m - Y0 - v - h - E1 - k - a:1 - E:1 - G - f - j - T - a1 - p - c - au:1 - i:1 - O:1 - I:1 - E0 - I1 - r_0 - t_h - k_h - Y1 - ei1 - i0 - ou:1 - ei:1 - u:1 - O1 - N - l_0 - '91' - ai0 - au1 - ou0 - n_0 - ei0 - ai:1 - O0 - ou1 - ai1 - i1 - '9:1' - '90' - au0 - x - c_h - 9i:1 - C - p_h - u0 - Y:1 - J - 9i1 - u1 - 9i0 - N_0 - m_0 - J_0 - Yi0 - Oi1 - Yi1 - Oi0 - au:0 - '9:0' - E:0 - <sos/eos> odim: null model_conf: {} use_preprocessor: true token_type: phn bpemodel: null non_linguistic_symbols: null cleaner: null g2p: null feats_extract: fbank feats_extract_conf: n_fft: 1024 hop_length: 256 win_length: null fs: 22050 fmin: 80 fmax: 7600 n_mels: 80 normalize: global_mvn normalize_conf: stats_file: exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/train/feats_stats.npz tts: fastspeech2 tts_conf: adim: 384 aheads: 2 elayers: 4 eunits: 1536 dlayers: 4 dunits: 1536 positionwise_layer_type: conv1d positionwise_conv_kernel_size: 3 duration_predictor_layers: 2 duration_predictor_chans: 256 duration_predictor_kernel_size: 3 postnet_layers: 5 postnet_filts: 5 postnet_chans: 256 use_masking: true use_scaled_pos_enc: true encoder_normalize_before: true decoder_normalize_before: true reduction_factor: 1 init_type: xavier_uniform init_enc_alpha: 1.0 init_dec_alpha: 1.0 transformer_enc_dropout_rate: 0.2 transformer_enc_positional_dropout_rate: 0.2 transformer_enc_attn_dropout_rate: 0.2 transformer_dec_dropout_rate: 0.2 transformer_dec_positional_dropout_rate: 0.2 transformer_dec_attn_dropout_rate: 0.2 pitch_predictor_layers: 5 pitch_predictor_chans: 256 pitch_predictor_kernel_size: 5 pitch_predictor_dropout: 0.5 pitch_embed_kernel_size: 1 pitch_embed_dropout: 0.0 stop_gradient_from_pitch_predictor: true energy_predictor_layers: 2 energy_predictor_chans: 256 energy_predictor_kernel_size: 3 energy_predictor_dropout: 0.5 energy_embed_kernel_size: 1 energy_embed_dropout: 0.0 stop_gradient_from_energy_predictor: false pitch_extract: dio pitch_extract_conf: fs: 22050 n_fft: 1024 hop_length: 256 f0max: 400 f0min: 80 reduction_factor: 1 pitch_normalize: global_mvn pitch_normalize_conf: stats_file: exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/train/pitch_stats.npz energy_extract: energy energy_extract_conf: fs: 22050 n_fft: 1024 hop_length: 256 win_length: null reduction_factor: 1 energy_normalize: global_mvn energy_normalize_conf: stats_file: exp/a/tts_train_tacotron2_raw_phn_none/decode_use_teacher_forcingtrue_train.loss.ave/stats/train/energy_stats.npz required: - output_dir - token_list version: 0.10.7a1 distributed: false ``` </details> ### Citing ESPnet ```BibTex @inproceedings{watanabe2018espnet, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, title={{ESPnet}: End-to-End Speech Processing Toolkit}, year={2018}, booktitle={Proceedings of Interspeech}, pages={2207--2211}, doi={10.21437/Interspeech.2018-1456}, url={http://dx.doi.org/10.21437/Interspeech.2018-1456} } @inproceedings{hayashi2020espnet, title={{Espnet-TTS}: Unified, reproducible, and integratable open source end-to-end text-to-speech toolkit}, author={Hayashi, Tomoki and Yamamoto, Ryuichi and Inoue, Katsuki and Yoshimura, Takenori and Watanabe, Shinji and Toda, Tomoki and Takeda, Kazuya and Zhang, Yu and Tan, Xu}, booktitle={Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, pages={7654--7658}, year={2020}, organization={IEEE} } ``` or arXiv: ```bibtex @misc{watanabe2018espnet, title={ESPnet: End-to-End Speech Processing Toolkit}, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, year={2018}, eprint={1804.00015}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```

7723afd5623652e29f2fe732cb153590

ChaoLi/xlm-roberta-base-finetuned-panx-de-fr

ChaoLi

xlm-roberta

10

5

transformers

0

token-classification

true

false

mit

null

0

['generated_from_trainer']

true

1,315

false

# xlm-roberta-base-finetuned-panx-de-fr This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1643 - F1: 0.8626 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 24 - eval_batch_size: 24 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:----:|:---------------:|:------:| | 0.2891 | 1.0 | 715 | 0.1780 | 0.8288 | | 0.1472 | 2.0 | 1430 | 0.1633 | 0.8488 | | 0.0948 | 3.0 | 2145 | 0.1643 | 0.8626 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.11.0 - Datasets 1.16.1 - Tokenizers 0.10.3

f9e5a17e5a3aa6e25910b2b3ae17da5e

testorg2/larger_fork

testorg2

bert

13

5

sentence-transformers

0

sentence-similarity

true

false

apache-2.0

['multilingual']

null

0

['sentence-transformers', 'feature-extraction', 'sentence-similarity', 'transformers']

false

true

3,613

false

# sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search. ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2') embeddings = model.encode(sentences) print(embeddings) ``` ## Usage (HuggingFace Transformers) Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. ```python from transformers import AutoTokenizer, AutoModel import torch #Mean Pooling - Take attention mask into account for correct averaging def mean_pooling(model_output, attention_mask): token_embeddings = model_output[0] #First element of model_output contains all token embeddings input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float() return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9) # Sentences we want sentence embeddings for sentences = ['This is an example sentence', 'Each sentence is converted'] # Load model from HuggingFace Hub tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2') model = AutoModel.from_pretrained('sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2') # Tokenize sentences encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt') # Compute token embeddings with torch.no_grad(): model_output = model(**encoded_input) # Perform pooling. In this case, max pooling. sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask']) print("Sentence embeddings:") print(sentence_embeddings) ``` ## Evaluation Results For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False}) ) ``` ## Citing & Authors This model was trained by [sentence-transformers](https://www.sbert.net/). If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084): ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "http://arxiv.org/abs/1908.10084", } ```

dbc62b5efb883f9702c1215685963b03

pidanr/bert-finetuned-race

pidanr

bert

12

1

transformers

0

multiple-choice

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,412

false

# bert-finetuned-race This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.3863 - Accuracy: 0.2982 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 1 - eval_batch_size: 1 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 8 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 1 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 1.3936 | 0.25 | 3100 | 1.3863 | 0.2418 | | 1.3768 | 0.51 | 6200 | 1.3863 | 0.2483 | | 1.3954 | 0.76 | 9300 | 1.3863 | 0.2982 | ### Framework versions - Transformers 4.19.1 - Pytorch 1.11.0+cu113 - Datasets 2.2.1 - Tokenizers 0.12.1

b06166a54868a423b46ba4b93ccbbde9

jcblaise/roberta-tagalog-large

jcblaise

roberta

10

607

transformers

0

fill-mask

true

false

cc-by-sa-4.0

['tl']

null

0

['roberta', 'tagalog', 'filipino']

false

true

1,152

false

# RoBERTa Tagalog Large Tagalog RoBERTa trained as an improvement over our previous Tagalog pretrained Transformers. Trained with TLUnified, a newer, larger, more topically-varied pretraining corpus for Filipino. This model is part of a larger research project. We open-source the model to allow greater usage within the Filipino NLP community. This model is a cased model. We do not release uncased RoBERTa models. ## Citations All model details and training setups can be found in our papers. If you use our model or find it useful in your projects, please cite our work: ``` @article{cruz2021improving, title={Improving Large-scale Language Models and Resources for Filipino}, author={Jan Christian Blaise Cruz and Charibeth Cheng}, journal={arXiv preprint arXiv:2111.06053}, year={2021} } ``` ## Data and Other Resources Data used to train this model as well as other benchmark datasets in Filipino can be found in my website at https://blaisecruz.com ## Contact If you have questions, concerns, or if you just want to chat about NLP and low-resource languages in general, you may reach me through my work email at me@blaisecruz.com

028f75600bbaf8862d953b2eee199b21

IDEA-CCNL/Yuyuan-GPT2-3.5B

IDEA-CCNL

gpt2

10

54

transformers

2

text-generation

true

false

apache-2.0

['en']

null

0

[]

false

true

2,891

false

# Yuyuan-GPT2-3.5B - Github: [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM) - Docs: [Fengshenbang-Docs](https://fengshenbang-doc.readthedocs.io/) ## 简介 Brief Introduction 目前最大的，医疗领域的生成语言模型GPT2。 The currently largest, generative language model GPT2 in the medical domain. ## 模型分类 Model Taxonomy | 需求 Demand | 任务 Task | 系列 Series | 模型 Model | 参数 Parameter | 额外 Extra | | :----: | :----: | :----: | :----: | :----: | :----: | | 特殊 Special | 领域 Domain | 余元 Yuyuan | GPT2 | 3.5B | - | ## 模型信息 Model Information 我们采用与Wenzhong-GPT2-3.5B相同的架构，在50GB的医学(PubMed)语料库上进行预训练。我们使用了32个NVIDIA A100显卡大约7天。我们的Yuyuan-GPT2-3.5B是医疗领域最大的开源的GPT2模型。进一步地，模型可以通过计算困惑度（PPL）来判断事实。为了完成问答功能，我们将短语模式从疑问的形式转换为了陈述句。 We adopt the same architecture as Wenzhong-GPT2-3.5B to be pre-trained on 50 GB medical (PubMed) corpus. We use 32 NVIDIA A100 GPUs for about 7 days. Our Yuyuan-GPT2-3.5B is the largest open-source GPT2 model in the medical domain. We further allow the model to judge facts by computing perplexity (PPL). To accomplish question-and-answer functionality, we transform the phrase pattern from interrogative to declarative. ## 使用 Usage ### 加载模型 Loading Models ```python from transformers import GPT2Tokenizer, GPT2Model tokenizer = GPT2Tokenizer.from_pretrained('IDEA-CCNL/Yuyuan-GPT2-3.5B') model = GPT2Model.from_pretrained('IDEA-CCNL/Yuyuan-GPT2-3.5B') text = "Replace me by any text you'd like." encoded_input = tokenizer(text, return_tensors='pt') output = model(**encoded_input) ``` ### 使用示例 Usage Examples ```python from transformers import pipeline, set_seed set_seed(55) generator = pipeline('text-generation', model='IDEA-CCNL/Yuyuan-GPT2-3.5B') generator("Diabetics should not eat", max_length=30, num_return_sequences=1) ``` ## 引用 Citation 如果您在您的工作中使用了我们的模型，可以引用我们的[论文](https://arxiv.org/abs/2209.02970)： If you are using the resource for your work, please cite the our [paper](https://arxiv.org/abs/2209.02970): ```text @article{fengshenbang, author = {Junjie Wang and Yuxiang Zhang and Lin Zhang and Ping Yang and Xinyu Gao and Ziwei Wu and Xiaoqun Dong and Junqing He and Jianheng Zhuo and Qi Yang and Yongfeng Huang and Xiayu Li and Yanghan Wu and Junyu Lu and Xinyu Zhu and Weifeng Chen and Ting Han and Kunhao Pan and Rui Wang and Hao Wang and Xiaojun Wu and Zhongshen Zeng and Chongpei Chen and Ruyi Gan and Jiaxing Zhang}, title = {Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence}, journal = {CoRR}, volume = {abs/2209.02970}, year = {2022} } ``` 也可以引用我们的[网站](https://github.com/IDEA-CCNL/Fengshenbang-LM/): You can also cite our [website](https://github.com/IDEA-CCNL/Fengshenbang-LM/): ```text @misc{Fengshenbang-LM, title={Fengshenbang-LM}, author={IDEA-CCNL}, year={2021}, howpublished={\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}}, } ```

864e2bae1a402fbe6b51b0f3d457d2f8

muhtasham/tiny-vanilla-target-glue-mnli

muhtasham

bert

10

1

transformers

0

text-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

2,515

false

# tiny-vanilla-target-glue-mnli This model is a fine-tuned version of [google/bert_uncased_L-2_H-128_A-2](https://huggingface.co/google/bert_uncased_L-2_H-128_A-2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8100 - Accuracy: 0.6375 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: constant - num_epochs: 200 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:-----:|:---------------:|:--------:| | 1.0866 | 0.04 | 500 | 1.0515 | 0.4557 | | 1.0101 | 0.08 | 1000 | 0.9526 | 0.5612 | | 0.9599 | 0.12 | 1500 | 0.9195 | 0.5802 | | 0.9378 | 0.16 | 2000 | 0.9018 | 0.5930 | | 0.9229 | 0.2 | 2500 | 0.8904 | 0.5954 | | 0.9182 | 0.24 | 3000 | 0.8802 | 0.6033 | | 0.9019 | 0.29 | 3500 | 0.8738 | 0.6070 | | 0.8971 | 0.33 | 4000 | 0.8613 | 0.6154 | | 0.8788 | 0.37 | 4500 | 0.8593 | 0.6172 | | 0.8856 | 0.41 | 5000 | 0.8508 | 0.6194 | | 0.8751 | 0.45 | 5500 | 0.8404 | 0.6256 | | 0.8718 | 0.49 | 6000 | 0.8445 | 0.6248 | | 0.8739 | 0.53 | 6500 | 0.8333 | 0.6306 | | 0.8653 | 0.57 | 7000 | 0.8363 | 0.6280 | | 0.8588 | 0.61 | 7500 | 0.8213 | 0.6376 | | 0.8587 | 0.65 | 8000 | 0.8215 | 0.6360 | | 0.8544 | 0.69 | 8500 | 0.8268 | 0.6292 | | 0.8556 | 0.73 | 9000 | 0.8045 | 0.6463 | | 0.8445 | 0.77 | 9500 | 0.8187 | 0.6328 | | 0.836 | 0.81 | 10000 | 0.8021 | 0.6446 | | 0.8399 | 0.86 | 10500 | 0.8100 | 0.6375 | ### Framework versions - Transformers 4.26.0.dev0 - Pytorch 1.13.0+cu116 - Datasets 2.8.1.dev0 - Tokenizers 0.13.2

4c861b68514674884df19079004615a4

BlackKakapo/t5-base-paraphrase-ro-v2

BlackKakapo

t5

8

4

transformers

0

text2text-generation

true

false

['apache-2.0']

['ro']

null

0

[]

false

true

1,987

false

# Romanian paraphrase ![v2.0](https://img.shields.io/badge/V.2-19.08.2022-brightgreen) Fine-tune t5-base-paraphrase-ro model for paraphrase. Since there is no Romanian dataset for paraphrasing, I had to create my own [dataset](https://huggingface.co/datasets/BlackKakapo/paraphrase-ro-v2). The dataset contains ~30k examples. ### How to use ```python from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("BlackKakapo/t5-base-paraphrase-ro-v2") model = AutoModelForSeq2SeqLM.from_pretrained("BlackKakapo/t5-base-paraphrase-ro-v2") ``` ### Or ```python from transformers import T5ForConditionalGeneration, T5TokenizerFast model = T5ForConditionalGeneration.from_pretrained("BlackKakapo/t5-base-paraphrase-ro-v2") tokenizer = T5TokenizerFast.from_pretrained("BlackKakapo/t5-base-paraphrase-ro-v2") ``` ### Generate ```python text = "Într-un interviu pentru Radio Europa Liberă România, acesta a menționat că Bucureștiul este pregătit oricând și ar dura doar o oră de la solicitare, până când gazele ar ajunge la Chișinău." encoding = tokenizer.encode_plus(text, pad_to_max_length=True, return_tensors="pt") input_ids, attention_masks = encoding["input_ids"], encoding["attention_mask"] beam_outputs = model.generate( input_ids=input_ids, attention_mask=attention_masks, do_sample=True, max_length=256, top_k=20, top_p=0.9, early_stopping=False, num_return_sequences=5 ) final_outputs = [] for beam_output in beam_outputs: text_para = tokenizer.decode(beam_output, skip_special_tokens=True,clean_up_tokenization_spaces=True) if text.lower() != text_para.lower() or text not in final_outputs: final_outputs.append(text_para) print(final_outputs) ``` ### Output ```out ['Într-un interviu cu Radio Europa Liberă România, el a spus că Bucureștiul este pregătit în orice moment și ar dura doar o oră de la cererea până când gazele ar ajunge la Chișinău.'] ```

37d1c681f4f33dd9895a65ba57c395ed

ix502iv/wa2vec2-large-xls-r-colab_turkish

ix502iv

wav2vec2

12

25

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

['common_voice']

null

0

['generated_from_trainer']

true

1,784

false

# wa2vec2-large-xls-r-colab_turkish This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset. It achieves the following results on the evaluation set: - Loss: 0.3941 - Wer: 0.3812 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 4.0265 | 3.67 | 400 | 0.7368 | 0.8192 | | 0.4253 | 7.34 | 800 | 0.4467 | 0.5111 | | 0.1902 | 11.01 | 1200 | 0.4423 | 0.4723 | | 0.1293 | 14.68 | 1600 | 0.3854 | 0.4216 | | 0.0989 | 18.35 | 2000 | 0.3997 | 0.4197 | | 0.0745 | 22.02 | 2400 | 0.4133 | 0.4182 | | 0.0598 | 25.69 | 2800 | 0.3962 | 0.3925 | | 0.0488 | 29.36 | 3200 | 0.3941 | 0.3812 | ### Framework versions - Transformers 4.25.1 - Pytorch 1.13.0+cu116 - Datasets 2.7.1 - Tokenizers 0.13.2

5e9db94a8ff8a95838e6deda6a985a7c

sanamoin/wav2vec2-base-timit-demo-google-colab

sanamoin

wav2vec2

12

5

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,021

false

# wav2vec2-base-timit-demo-google-colab This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results ### Framework versions - Transformers 4.17.0 - Pytorch 1.11.0+cu113 - Datasets 1.18.3 - Tokenizers 0.12.1

6bbbd2d3ace5de75d9922b538b1599a6

cammy/bart-large-cnn-finetuned-weaksup-1000-pad-early-new1

cammy

bart

11

1

transformers

0

text2text-generation

true

false

mit

null

0

['generated_from_trainer']

true

1,575

false

# bart-large-cnn-finetuned-weaksup-1000-pad-early-new1 This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.4948 - Rouge1: 28.1465 - Rouge2: 13.4076 - Rougel: 22.2763 - Rougelsum: 25.2087 - Gen Len: 68.58 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 1 - eval_batch_size: 1 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 10 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:| | 0.156 | 1.0 | 1000 | 0.4377 | 27.8782 | 13.1274 | 21.2329 | 24.6465 | 66.25 | | 0.0843 | 2.0 | 2000 | 0.4948 | 28.1465 | 13.4076 | 22.2763 | 25.2087 | 68.58 | ### Framework versions - Transformers 4.16.2 - Pytorch 1.10.2 - Datasets 1.18.3 - Tokenizers 0.11.0

7c4c05d74542b6c59d40c781fd6803aa

KoichiYasuoka/roberta-large-korean-ud-goeswith

KoichiYasuoka

roberta

10

8

transformers

1

token-classification

true

false

cc-by-sa-4.0

['ko']

['universal_dependencies']

null

0

['korean', 'token-classification', 'pos', 'dependency-parsing']

false

true

2,722

false

# roberta-large-korean-ud-goeswith ## Model Description This is a RoBERTa model pre-trained on Korean texts for POS-tagging and dependency-parsing (using `goeswith` for subwords), derived from [roberta-large-korean-hanja](https://huggingface.co/KoichiYasuoka/roberta-large-korean-hanja). ## How to Use ```py class UDgoeswith(object): def __init__(self,bert): from transformers import AutoTokenizer,AutoModelForTokenClassification self.tokenizer=AutoTokenizer.from_pretrained(bert) self.model=AutoModelForTokenClassification.from_pretrained(bert) def __call__(self,text): import numpy,torch,ufal.chu_liu_edmonds w=self.tokenizer(text,return_offsets_mapping=True) v=w["input_ids"] x=[v[0:i]+[self.tokenizer.mask_token_id]+v[i+1:]+[j] for i,j in enumerate(v[1:-1],1)] with torch.no_grad(): e=self.model(input_ids=torch.tensor(x)).logits.numpy()[:,1:-2,:] r=[1 if i==0 else -1 if j.endswith("|root") else 0 for i,j in sorted(self.model.config.id2label.items())] e+=numpy.where(numpy.add.outer(numpy.identity(e.shape[0]),r)==0,0,numpy.nan) g=self.model.config.label2id["X|_|goeswith"] r=numpy.tri(e.shape[0]) for i in range(e.shape[0]): for j in range(i+2,e.shape[1]): r[i,j]=r[i,j-1] if numpy.nanargmax(e[i,j-1])==g else 1 e[:,:,g]+=numpy.where(r==0,0,numpy.nan) m=numpy.full((e.shape[0]+1,e.shape[1]+1),numpy.nan) m[1:,1:]=numpy.nanmax(e,axis=2).transpose() p=numpy.zeros(m.shape) p[1:,1:]=numpy.nanargmax(e,axis=2).transpose() for i in range(1,m.shape[0]): m[i,0],m[i,i],p[i,0]=m[i,i],numpy.nan,p[i,i] h=ufal.chu_liu_edmonds.chu_liu_edmonds(m)[0] if [0 for i in h if i==0]!=[0]: m[:,0]+=numpy.where(m[:,0]==numpy.nanmax(m[[i for i,j in enumerate(h) if j==0],0]),0,numpy.nan) m[[i for i,j in enumerate(h) if j==0]]+=[0 if i==0 or j==0 else numpy.nan for i,j in enumerate(h)] h=ufal.chu_liu_edmonds.chu_liu_edmonds(m)[0] u="# text = "+text+"\n" v=[(s,e) for s,e in w["offset_mapping"] if s<e] for i,(s,e) in enumerate(v,1): q=self.model.config.id2label[p[i,h[i]]].split("|") u+="\t".join([str(i),text[s:e],"_",q[0],"_","|".join(q[1:-1]),str(h[i]),q[-1],"_","_" if i<len(v) and e<v[i][0] else "SpaceAfter=No"])+"\n" return u+"\n" nlp=UDgoeswith("KoichiYasuoka/roberta-large-korean-ud-goeswith") print(nlp("홍시 맛이 나서 홍시라 생각한다.")) ``` with [ufal.chu-liu-edmonds](https://pypi.org/project/ufal.chu-liu-edmonds/). Or without ufal.chu-liu-edmonds: ``` from transformers import pipeline nlp=pipeline("universal-dependencies","KoichiYasuoka/roberta-large-korean-ud-goeswith",trust_remote_code=True,aggregation_strategy="simple") print(nlp("홍시 맛이 나서 홍시라 생각한다.")) ```

bb016840147e49a875d083cd8709533d

birgermoell/wav2vec2-common_voice-tr-demo

birgermoell

wav2vec2

15

7

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['sv-SE']

['common_voice']

null

0

['automatic-speech-recognition', 'common_voice', 'generated_from_trainer']

true

2,522

false

# wav2vec2-common_voice-tr-demo This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the COMMON_VOICE - SV-SE dataset. It achieves the following results on the evaluation set: - Loss: 0.5528 - Wer: 0.3811 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 15.0 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | No log | 0.74 | 100 | 3.4444 | 1.0 | | No log | 1.47 | 200 | 2.9421 | 1.0 | | No log | 2.21 | 300 | 2.2802 | 1.0137 | | No log | 2.94 | 400 | 0.9683 | 0.7611 | | 3.7264 | 3.68 | 500 | 0.7941 | 0.6594 | | 3.7264 | 4.41 | 600 | 0.6695 | 0.5751 | | 3.7264 | 5.15 | 700 | 0.6507 | 0.5314 | | 3.7264 | 5.88 | 800 | 0.5731 | 0.4927 | | 3.7264 | 6.62 | 900 | 0.5723 | 0.4580 | | 0.4592 | 7.35 | 1000 | 0.5913 | 0.4479 | | 0.4592 | 8.09 | 1100 | 0.5562 | 0.4423 | | 0.4592 | 8.82 | 1200 | 0.5566 | 0.4292 | | 0.4592 | 9.56 | 1300 | 0.5492 | 0.4303 | | 0.4592 | 10.29 | 1400 | 0.5665 | 0.4331 | | 0.2121 | 11.03 | 1500 | 0.5610 | 0.4084 | | 0.2121 | 11.76 | 1600 | 0.5703 | 0.4014 | | 0.2121 | 12.5 | 1700 | 0.5669 | 0.3898 | | 0.2121 | 13.24 | 1800 | 0.5586 | 0.3962 | | 0.2121 | 13.97 | 1900 | 0.5656 | 0.3897 | | 0.1326 | 14.71 | 2000 | 0.5565 | 0.3813 | ### Framework versions - Transformers 4.16.0.dev0 - Pytorch 1.10.1+cu113 - Datasets 1.18.0 - Tokenizers 0.10.3

9e0f37c6898c02c1d9ef79532020a84c

pinot/wav2vec2-base-timit-demo-colab

pinot

wav2vec2

12

8

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,641

false

# wav2vec2-base-timit-demo-colab This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4548 - Wer: 0.3373 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 32 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 3.3291 | 4.0 | 500 | 1.0403 | 0.7174 | | 0.5336 | 8.0 | 1000 | 0.4744 | 0.4489 | | 0.2155 | 12.0 | 1500 | 0.4476 | 0.3832 | | 0.1256 | 16.0 | 2000 | 0.4358 | 0.3639 | | 0.0867 | 20.0 | 2500 | 0.4634 | 0.3527 | | 0.0608 | 24.0 | 3000 | 0.4784 | 0.3466 | | 0.0476 | 28.0 | 3500 | 0.4548 | 0.3373 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.10.0+cu111 - Datasets 1.18.3 - Tokenizers 0.10.3

e68fc0967fc17c1a5aab948986bc0b7e

jo-kwsm/xlm-roberta-base-finetuned-panx-de

jo-kwsm

xlm-roberta

16

5

transformers

0

token-classification

true

false

mit

null

['xtreme']

null

0

['generated_from_trainer']

true

1,320

false

# xlm-roberta-base-finetuned-panx-de This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the xtreme dataset. It achieves the following results on the evaluation set: - Loss: 0.1408 - F1: 0.8646 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 24 - eval_batch_size: 24 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:----:|:---------------:|:------:| | 0.2626 | 1.0 | 525 | 0.1807 | 0.8067 | | 0.1307 | 2.0 | 1050 | 0.1388 | 0.8526 | | 0.0829 | 3.0 | 1575 | 0.1408 | 0.8646 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.12.1+cu102 - Datasets 1.16.1 - Tokenizers 0.10.3

18bff117b5d84ca0dccc9dd21c02231f

MultiBertGunjanPatrick/multiberts-seed-4-1900k

MultiBertGunjanPatrick

bert

7

4

transformers

0

null

true

false

apache-2.0

['en']

['bookcorpus', 'wikipedia']

null

0

['exbert', 'multiberts', 'multiberts-seed-4']

false

true

6,487

false

# MultiBERTs Seed 4 Checkpoint 1900k (uncased) Seed 4 intermediate checkpoint 1900k MultiBERTs (pretrained BERT) model on English language using a masked language modeling (MLM) objective. It was introduced in [this paper](https://arxiv.org/pdf/2106.16163.pdf) and first released in [this repository](https://github.com/google-research/language/tree/master/language/multiberts). This is an intermediate checkpoint. The final checkpoint can be found at [multiberts-seed-4](https://hf.co/multberts-seed-4). This model is uncased: it does not make a difference between english and English. Disclaimer: The team releasing MultiBERTs did not write a model card for this model so this model card has been written by [gchhablani](https://hf.co/gchhablani). ## Model description MultiBERTs models are transformers model pretrained on a large corpus of English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was pretrained with two objectives: - Masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then run the entire masked sentence through the model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like GPT which internally mask the future tokens. It allows the model to learn a bidirectional representation of the sentence. - Next sentence prediction (NSP): the models concatenates two masked sentences as inputs during pretraining. Sometimes they correspond to sentences that were next to each other in the original text, sometimes not. The model then has to predict if the two sentences were following each other or not. This way, the model learns an inner representation of the English language that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard classifier using the features produced by the MultiBERTs model as inputs. ## Intended uses & limitations You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=multiberts) to look for fine-tuned versions on a task that interests you. Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at model like GPT2. ### How to use Here is how to use this model to get the features of a given text in PyTorch: ```python from transformers import BertTokenizer, BertModel tokenizer = BertTokenizer.from_pretrained('multiberts-seed-4-1900k') model = BertModel.from_pretrained("multiberts-seed-4-1900k") text = "Replace me by any text you'd like." encoded_input = tokenizer(text, return_tensors='pt') output = model(**encoded_input) ``` ### Limitations and bias Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. This bias will also affect all fine-tuned versions of this model. For an understanding of bias of this particular checkpoint, please try out this checkpoint with the snippet present in the [Limitation and bias section](https://huggingface.co/bert-base-uncased#limitations-and-bias) of the [bert-base-uncased](https://huggingface.co/bert-base-uncased) checkpoint. ## Training data The MultiBERTs models were pretrained on [BookCorpus](https://yknzhu.wixsite.com/mbweb), a dataset consisting of 11,038 unpublished books and [English Wikipedia](https://en.wikipedia.org/wiki/English_Wikipedia) (excluding lists, tables and headers). ## Training procedure ### Preprocessing The texts are lowercased and tokenized using WordPiece and a vocabulary size of 30,000. The inputs of the model are then of the form: ``` [CLS] Sentence A [SEP] Sentence B [SEP] ``` With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two "sentences" has a combined length of less than 512 tokens. The details of the masking procedure for each sentence are the following: - 15% of the tokens are masked. - In 80% of the cases, the masked tokens are replaced by `[MASK]`. - In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace. - In the 10% remaining cases, the masked tokens are left as is. ### Pretraining The full model was trained on 16 Cloud TPU v2 chips for two million steps with a batch size of 256. The sequence length was set to 512 throughout. The optimizer used is Adam with a learning rate of 1e-4, \$\beta_{1} = 0.9\$ and \$\beta_{2} = 0.999\$, a weight decay of 0.01, learning rate warmup for 10,000 steps and linear decay of the learning rate after. ### BibTeX entry and citation info ```bibtex @article{DBLP:journals/corr/abs-2106-16163, author = {Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick}, title = {The MultiBERTs: {BERT} Reproductions for Robustness Analysis}, journal = {CoRR}, volume = {abs/2106.16163}, year = {2021}, url = {https://arxiv.org/abs/2106.16163}, eprinttype = {arXiv}, eprint = {2106.16163}, timestamp = {Mon, 05 Jul 2021 15:15:50 +0200}, biburl = {https://dblp.org/rec/journals/corr/abs-2106-16163.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} } ``` <a href="https://huggingface.co/exbert/?model=multiberts"> <img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png"> </a>

96950581ba05cf9577ab9502817298d8

shimdx/wav2vec2-base-demo-sagemaker

shimdx

wav2vec2

11

7

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,633

false

# wav2vec2-base-demo-sagemaker This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4713 - Wer: 0.3381 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 32 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 3.4274 | 4.0 | 500 | 1.2279 | 0.8902 | | 0.5778 | 8.0 | 1000 | 0.4838 | 0.4488 | | 0.2244 | 12.0 | 1500 | 0.4813 | 0.3793 | | 0.1299 | 16.0 | 2000 | 0.4878 | 0.3714 | | 0.0871 | 20.0 | 2500 | 0.4796 | 0.3539 | | 0.0635 | 24.0 | 3000 | 0.4554 | 0.3427 | | 0.0495 | 28.0 | 3500 | 0.4713 | 0.3381 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.10.0 - Datasets 1.14.0 - Tokenizers 0.10.3

463db7723dc1676d108ff4ea64e927d4

nickprock/bert-finetuned-ner-ontonotes

nickprock

bert

12

17

transformers

0

token-classification

true

false

apache-2.0

null

['ontonotes5']

null

0

['generated_from_trainer']

true

1,813

false

# bert-finetuned-ner-ontonotes This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the ontonotes5 dataset. It achieves the following results on the evaluation set: - Loss: 0.1503 - Precision: 0.8567 - Recall: 0.8842 - F1: 0.8702 - Accuracy: 0.9755 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 6 ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| | 0.0842 | 1.0 | 7491 | 0.0950 | 0.8524 | 0.8715 | 0.8618 | 0.9745 | | 0.0523 | 2.0 | 14982 | 0.1044 | 0.8449 | 0.8827 | 0.8634 | 0.9744 | | 0.036 | 3.0 | 22473 | 0.1118 | 0.8529 | 0.8843 | 0.8683 | 0.9760 | | 0.0231 | 4.0 | 29964 | 0.1240 | 0.8589 | 0.8805 | 0.8696 | 0.9752 | | 0.0118 | 5.0 | 37455 | 0.1416 | 0.8570 | 0.8804 | 0.8685 | 0.9753 | | 0.0077 | 6.0 | 44946 | 0.1503 | 0.8567 | 0.8842 | 0.8702 | 0.9755 | ### Framework versions - Transformers 4.22.1 - Pytorch 1.12.1+cu113 - Datasets 2.5.1 - Tokenizers 0.12.1

b26ee5f9a4e771b2f9bdd1b601782d2c

tftgregrge/mpid-bkdbj

tftgregrge

null

18

7

diffusers

0

text-to-image

false

creativeml-openrail-m

null

1

0

['text-to-image', 'stable-diffusion']

false

true

422

false

### mpid-bkdbj Dreambooth model trained by tftgregrge with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Sample pictures of this concept:

1ec17798b50ba763b86504dee863818c

mrm8488/convbert-base-spanish

mrm8488

convbert

9

13

transformers

1

feature-extraction

true

false

mit

['es']

['large_spanish_corpus']

null

0

[]

false

true

918

false

# ConvBERT base pre-trained on large_spanish_corpus The ConvBERT architecture is presented in the ["ConvBERT: Improving BERT with Span-based Dynamic Convolution"](https://arxiv.org/abs/2008.02496) paper. ## Metrics on evaluation set ``` disc_accuracy = 0.9488542 disc_auc = 0.8833056 disc_loss = 0.15933733 disc_precision = 0.79224133 disc_recall = 0.27443287 global_step = 1000000 loss = 9.658503 masked_lm_accuracy = 0.6177698 masked_lm_loss = 1.7050561 sampled_masked_lm_accuracy = 0.5379228 ``` ## Usage ```python from transformers import AutoModel, AutoTokenizer model_name = "mrm8488/convbert-base-spanish" tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModel.from_pretrained(model_name) ``` > Created by [Manuel Romero/@mrm8488](https://twitter.com/mrm8488) with the support of [Narrativa](https://www.narrativa.com/) > Made with <span style="color: #e25555;">&hearts;</span> in Spain

4f5671ffc8175e2f5a5abfea0ee6ca2c

anuragshas/whisper-small-mr

anuragshas

whisper

17

9

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['mr']

['mozilla-foundation/common_voice_11_0']

null

0

['whisper-event', 'generated_from_trainer']

true

459

false

# Whisper Small Marathi This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the mozilla-foundation/common_voice_11_0 mr dataset. It achieves the following results on the evaluation set: - Loss: 0.4888 - Wer: 19.71

add722c0967c0ac6d976d81b21ef5019

jjglilleberg/xlm-roberta-base-finetuned-panx-de

jjglilleberg

xlm-roberta

11

4

transformers

0

token-classification

true

false

mit

null

['xtreme']

null

0

['generated_from_trainer']

true

1,418

false

# xlm-roberta-base-finetuned-panx-de This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the xtreme dataset. It achieves the following results on the evaluation set: - Loss: 0.1518 - F1: 0.8616 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 16 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:----:|:---------------:|:------:| | No log | 1.0 | 786 | 0.1926 | 0.8138 | | No log | 2.0 | 1572 | 0.1580 | 0.8493 | | No log | 3.0 | 2358 | 0.1518 | 0.8616 | ### Framework versions - Transformers 4.25.1 - Pytorch 1.13.0+cu116 - Datasets 2.8.0 - Tokenizers 0.13.2

ef68d9fff73a564a3777dd6f405671e8

yanaiela/roberta-base-epoch_5

yanaiela

roberta

9

3

transformers

0

fill-mask

true

false

mit

['en']

['wikipedia', 'bookcorpus']

null

0

['roberta-base', 'roberta-base-epoch_5']

false

true

2,100

false

# RoBERTa, Intermediate Checkpoint - Epoch 5 This model is part of our reimplementation of the [RoBERTa model](https://arxiv.org/abs/1907.11692), trained on Wikipedia and the Book Corpus only. We train this model for almost 100K steps, corresponding to 83 epochs. We provide the 84 checkpoints (including the randomly initialized weights before the training) to provide the ability to study the training dynamics of such models, and other possible use-cases. These models were trained in part of a work that studies how simple statistics from data, such as co-occurrences affects model predictions, which are described in the paper [Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions](https://arxiv.org/abs/2207.14251). This is RoBERTa-base epoch_5. ## Model Description This model was captured during a reproduction of [RoBERTa-base](https://huggingface.co/roberta-base), for English: it is a Transformers model pretrained on a large corpus of English data, using the Masked Language Modelling (MLM). The intended uses, limitations, training data and training procedure for the fully trained model are similar to [RoBERTa-base](https://huggingface.co/roberta-base). Two major differences with the original model: * We trained our model for 100K steps, instead of 500K * We only use Wikipedia and the Book Corpus, as corpora which are publicly available. ### How to use Using code from [RoBERTa-base](https://huggingface.co/roberta-base), here is an example based on PyTorch: ``` from transformers import pipeline model = pipeline("fill-mask", model='yanaiela/roberta-base-epoch_83', device=-1, top_k=10) model("Hello, I'm the <mask> RoBERTa-base language model") ``` ## Citation info ```bibtex @article{2207.14251, Author = {Yanai Elazar and Nora Kassner and Shauli Ravfogel and Amir Feder and Abhilasha Ravichander and Marius Mosbach and Yonatan Belinkov and Hinrich Schütze and Yoav Goldberg}, Title = {Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions}, Year = {2022}, Eprint = {arXiv:2207.14251}, } ```

05d0b9edcd8c0e5461d373a738d7248b

gojiteji/text2QR

gojiteji

null

4

0

null

0

null

false

odbl

null

0

[]

false

true

649

false

This is a diffusion model fine-tuned with [QRsst2](https://huggingface.co/datasets/gojiteji/QRsst2). This model generates a QR code from text. Please clone this repository and replace [LambdaLabsML's example's inference code ](https://github.com/LambdaLabsML/examples/blob/767e1101b0125202871812ec7e1b5c46aa9c8d95/stable-diffusion-finetuning/pokemon_finetune.ipynb). checkpoint filename check pint name with `main.ckpt`. The below images are examples of an input :`The way to get started is to quit talking and begin doing.` ![example QR codes](example.png) sample code is here:https://github.com/gojiteji/text2QR/blob/main/samplecode.ipynb

af3a6a7c847216cc5e1f0dfc491a3638

Helsinki-NLP/opus-mt-de-lt

Helsinki-NLP

marian

10

16

transformers

0

translation

true

false

apache-2.0

null

0

['translation']

false

true

770

false

### opus-mt-de-lt * source languages: de * target languages: lt * OPUS readme: [de-lt](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/de-lt/README.md) * dataset: opus * model: transformer-align * pre-processing: normalization + SentencePiece * download original weights: [opus-2020-01-20.zip](https://object.pouta.csc.fi/OPUS-MT-models/de-lt/opus-2020-01-20.zip) * test set translations: [opus-2020-01-20.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/de-lt/opus-2020-01-20.test.txt) * test set scores: [opus-2020-01-20.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/de-lt/opus-2020-01-20.eval.txt) ## Benchmarks | testset | BLEU | chr-F | |-----------------------|-------|-------| | Tatoeba.de.lt | 37.9 | 0.633 |

f3b94b7c54986c506cf08cd2ccf46e26

jonatasgrosman/exp_w2v2r_es_xls-r_age_teens-8_sixties-2_s287

jonatasgrosman

wav2vec2

10

0

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['es']

['mozilla-foundation/common_voice_7_0']

null

0

['automatic-speech-recognition', 'es']

false

true

475

false

# exp_w2v2r_es_xls-r_age_teens-8_sixties-2_s287 Fine-tuned [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) for speech recognition using the train split of [Common Voice 7.0 (es)](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0). When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool.

9e4fa07b9e92cae18fe6ebb83b38f6b7

D3xter1922/electra-base-discriminator-finetuned-cola

D3xter1922

electra

13

4

transformers

0

text-classification

true

false

apache-2.0

null

['glue']

null

0

['generated_from_trainer']

true

1,595

false

# electra-base-discriminator-finetuned-cola This model is a fine-tuned version of [google/electra-base-discriminator](https://huggingface.co/google/electra-base-discriminator) on the glue dataset. It achieves the following results on the evaluation set: - Loss: 0.6367 - Matthews Correlation: 0.6824 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation | |:-------------:|:-----:|:----:|:---------------:|:--------------------:| | 0.4139 | 1.0 | 535 | 0.4137 | 0.6381 | | 0.2452 | 2.0 | 1070 | 0.4887 | 0.6504 | | 0.17 | 3.0 | 1605 | 0.5335 | 0.6757 | | 0.1135 | 4.0 | 2140 | 0.6367 | 0.6824 | | 0.0817 | 5.0 | 2675 | 0.6742 | 0.6755 | ### Framework versions - Transformers 4.15.0 - Pytorch 1.10.0+cu111 - Datasets 1.17.0 - Tokenizers 0.10.3

e4458928f35e11e6c14619cc18fe8be9

pyf98/chime4_e_branchformer_e10

pyf98

null

33

5

espnet

0

automatic-speech-recognition

false

cc-by-4.0

['en']

['chime4']

null

0

['espnet', 'audio', 'automatic-speech-recognition']

false

true

8,780

false

## ESPnet2 ASR model ### `pyf98/chime4_e_branchformer_e10` This model was trained by Yifan Peng using chime4 recipe in [espnet](https://github.com/espnet/espnet/). References: - [E-Branchformer: Branchformer with Enhanced merging for speech recognition (SLT 2022)](https://arxiv.org/abs/2210.00077) - [Branchformer: Parallel MLP-Attention Architectures to Capture Local and Global Context for Speech Recognition and Understanding (ICML 2022)](https://proceedings.mlr.press/v162/peng22a.html) ### Demo: How to use in ESPnet2 Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html) if you haven't done that already. ```bash cd espnet git checkout ad91279f0108d54bd22abe29671b376f048822c5 pip install -e . cd egs2/chime4/asr1 ./run.sh --skip_data_prep false --skip_train true --download_model pyf98/chime4_e_branchformer_e10 ```  # RESULTS ## Environments - date: `Wed Dec 28 15:49:24 EST 2022` - python version: `3.9.15 (main, Nov 24 2022, 14:31:59) [GCC 11.2.0]` - espnet version: `espnet 202211` - pytorch version: `pytorch 1.12.1` - Git hash: `f9a8009aef6ff9ba192a78c19b619ae4a9f3b9d2` - Commit date: `Wed Dec 28 00:30:54 2022 -0500` ## asr_train_asr_e_branchformer_e10_mlp1024_linear1024_macaron_lr1e-3_warmup25k_raw_en_char_sp ### WER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |---|---|---|---|---|---|---|---|---| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/dt05_real_beamformit_5mics|1640|27119|93.7|5.0|1.2|0.6|6.8|52.5| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/dt05_simu_beamformit_5mics|1640|27120|92.4|6.1|1.6|0.7|8.4|58.2| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/et05_real_beamformit_5mics|1320|21409|90.2|8.0|1.8|1.0|10.8|60.2| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/et05_simu_beamformit_5mics|1320|21416|88.4|9.3|2.4|1.4|13.0|66.1| ### CER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |---|---|---|---|---|---|---|---|---| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/dt05_real_beamformit_5mics|1640|160390|97.4|1.3|1.3|0.7|3.3|52.5| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/dt05_simu_beamformit_5mics|1640|160400|96.6|1.8|1.7|0.9|4.3|58.2| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/et05_real_beamformit_5mics|1320|126796|95.7|2.3|2.0|1.1|5.4|60.2| |decode_asr_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/et05_simu_beamformit_5mics|1320|126812|94.4|2.8|2.8|1.5|7.2|66.1| ### TER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |---|---|---|---|---|---|---|---|---| ## ASR config <details><summary>expand</summary> ``` config: conf/tuning/train_asr_e_branchformer_e10_mlp1024_linear1024_macaron_lr1e-3_warmup25k.yaml print_config: false log_level: INFO dry_run: false iterator_type: sequence output_dir: exp/asr_train_asr_e_branchformer_e10_mlp1024_linear1024_macaron_lr1e-3_warmup25k_raw_en_char_sp ngpu: 1 seed: 2022 num_workers: 4 num_att_plot: 3 dist_backend: nccl dist_init_method: env:// dist_world_size: 2 dist_rank: 0 local_rank: 0 dist_master_addr: localhost dist_master_port: 33561 dist_launcher: null multiprocessing_distributed: true unused_parameters: false sharded_ddp: false cudnn_enabled: true cudnn_benchmark: false cudnn_deterministic: true collect_stats: false write_collected_feats: false max_epoch: 50 patience: null val_scheduler_criterion: - valid - loss early_stopping_criterion: - valid - loss - min best_model_criterion: - - valid - acc - max keep_nbest_models: 10 nbest_averaging_interval: 0 grad_clip: 5.0 grad_clip_type: 2.0 grad_noise: false accum_grad: 1 no_forward_run: false resume: true train_dtype: float32 use_amp: true log_interval: null use_matplotlib: true use_tensorboard: true create_graph_in_tensorboard: false use_wandb: false wandb_project: null wandb_id: null wandb_entity: null wandb_name: null wandb_model_log_interval: -1 detect_anomaly: false pretrain_path: null init_param: [] ignore_init_mismatch: false freeze_param: [] num_iters_per_epoch: null batch_size: 20 valid_batch_size: null batch_bins: 15000000 valid_batch_bins: null train_shape_file: - exp/asr_stats_raw_en_char_sp/train/speech_shape - exp/asr_stats_raw_en_char_sp/train/text_shape.char valid_shape_file: - exp/asr_stats_raw_en_char_sp/valid/speech_shape - exp/asr_stats_raw_en_char_sp/valid/text_shape.char batch_type: numel valid_batch_type: null fold_length: - 80000 - 150 sort_in_batch: descending sort_batch: descending multiple_iterator: false chunk_length: 500 chunk_shift_ratio: 0.5 num_cache_chunks: 1024 train_data_path_and_name_and_type: - - dump/raw/tr05_multi_noisy_si284_sp/wav.scp - speech - kaldi_ark - - dump/raw/tr05_multi_noisy_si284_sp/text - text - text valid_data_path_and_name_and_type: - - dump/raw/dt05_multi_isolated_1ch_track/wav.scp - speech - kaldi_ark - - dump/raw/dt05_multi_isolated_1ch_track/text - text - text allow_variable_data_keys: false max_cache_size: 0.0 max_cache_fd: 32 valid_max_cache_size: null optim: adam optim_conf: lr: 0.001 weight_decay: 1.0e-06 scheduler: warmuplr scheduler_conf: warmup_steps: 25000 token_list: - <blank> - <unk> - <space> - E - T - A - N - I - O - S - R - H - L - D - C - U - M - P - F - G - Y - W - B - V - K - . - X - '''' - J - Q - Z - ',' - '-' - '"' - <NOISE> - '*' - ':' - ( - ) - '?' - '&' - ; - '!' - / - '{' - '}' - '1' - '2' - '0' - $ - '8' - '9' - '6' - '3' - '5' - '7' - '4' - '~' - '`' - _ - <*IN*> - <*MR.*> - \ - ^ - <sos/eos> init: null input_size: null ctc_conf: dropout_rate: 0.0 ctc_type: builtin reduce: true ignore_nan_grad: null zero_infinity: true joint_net_conf: null use_preprocessor: true token_type: char bpemodel: null non_linguistic_symbols: data/nlsyms.txt cleaner: null g2p: null speech_volume_normalize: null rir_scp: null rir_apply_prob: 1.0 noise_scp: null noise_apply_prob: 1.0 noise_db_range: '13_15' short_noise_thres: 0.5 frontend: default frontend_conf: n_fft: 512 win_length: 400 hop_length: 160 fs: 16k specaug: specaug specaug_conf: apply_time_warp: true time_warp_window: 5 time_warp_mode: bicubic apply_freq_mask: true freq_mask_width_range: - 0 - 27 num_freq_mask: 2 apply_time_mask: true time_mask_width_ratio_range: - 0.0 - 0.05 num_time_mask: 2 normalize: global_mvn normalize_conf: stats_file: exp/asr_stats_raw_en_char_sp/train/feats_stats.npz model: espnet model_conf: ctc_weight: 0.3 lsm_weight: 0.1 length_normalized_loss: false preencoder: null preencoder_conf: {} encoder: e_branchformer encoder_conf: output_size: 256 attention_heads: 4 attention_layer_type: rel_selfattn pos_enc_layer_type: rel_pos rel_pos_type: latest cgmlp_linear_units: 1024 cgmlp_conv_kernel: 31 use_linear_after_conv: false gate_activation: identity num_blocks: 10 dropout_rate: 0.1 positional_dropout_rate: 0.1 attention_dropout_rate: 0.1 input_layer: conv2d layer_drop_rate: 0.0 linear_units: 1024 positionwise_layer_type: linear use_ffn: true macaron_ffn: true merge_conv_kernel: 31 postencoder: null postencoder_conf: {} decoder: transformer decoder_conf: attention_heads: 4 linear_units: 2048 num_blocks: 6 dropout_rate: 0.1 positional_dropout_rate: 0.1 self_attention_dropout_rate: 0.1 src_attention_dropout_rate: 0.1 preprocessor: default preprocessor_conf: {} required: - output_dir - token_list version: '202211' distributed: true ``` </details> ### Citing ESPnet ```BibTex @inproceedings{watanabe2018espnet, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, title={{ESPnet}: End-to-End Speech Processing Toolkit}, year={2018}, booktitle={Proceedings of Interspeech}, pages={2207--2211}, doi={10.21437/Interspeech.2018-1456}, url={http://dx.doi.org/10.21437/Interspeech.2018-1456} } ``` or arXiv: ```bibtex @misc{watanabe2018espnet, title={ESPnet: End-to-End Speech Processing Toolkit}, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, year={2018}, eprint={1804.00015}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```

ecafb659f3e774136858327209a63a7a

charlemagne/distilbert-base-uncased-training-cola

charlemagne

distilbert

30

2

transformers

0

text-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,569

false

# distilbert-base-uncased-training-cola This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2215 - Matthews Correlation: 0.8777 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation | |:-------------:|:-----:|:----:|:---------------:|:--------------------:| | No log | 1.0 | 113 | 0.2954 | 0.7090 | | No log | 2.0 | 226 | 0.2212 | 0.8232 | | No log | 3.0 | 339 | 0.1899 | 0.8671 | | No log | 4.0 | 452 | 0.2006 | 0.8672 | | 0.19 | 5.0 | 565 | 0.2215 | 0.8777 | ### Framework versions - Transformers 4.17.0 - Pytorch 1.8.0+cu111 - Datasets 2.1.0 - Tokenizers 0.11.6

9e60e7a8909e99a2b7a70b421c3c4441

juierror/wav2vec2-large-xls-r-thai-test

juierror

wav2vec2

17

8

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

['common_voice']

null

0

['generated_from_trainer']

true

1,287

false

# wav2vec2-large-xls-r-thai-test This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the common_voice dataset. It achieves the following results on the evaluation set: - eval_loss: 0.7728 - eval_wer: 0.9490 - eval_runtime: 678.2819 - eval_samples_per_second: 3.226 - eval_steps_per_second: 0.404 - epoch: 2.56 - step: 600 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0003 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 16 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 400 - num_epochs: 5 - mixed_precision_training: Native AMP ### Framework versions - Transformers 4.15.0 - Pytorch 1.10.0+cu111 - Datasets 1.17.0 - Tokenizers 0.10.3

848947838da327187eab94513c9cb871

inkoziev/rugpt_interpreter

inkoziev

gpt2

12

14

transformers

5

text-generation

true

false

unlicense

['ru']

null

0

['PyTorch', 'Transformers', 'gpt2']

false

true

9,473

false

## Задача Incomplete Utterance Restoration Генеративная модель на основе [sberbank-ai/rugpt3large_based_on_gpt2](https://huggingface.co/sberbank-ai/rugpt3large_based_on_gpt2) для восстановления полного текста реплик в диалоге из контекста. Допустим, последние 2 строки диалога имеют вид: ``` - Как тебя зовут? - Джульетта Мао ``` Модель позволяет получить полный текст последней реплики, с раскрытыми анафорами, эллипсисами и т.д.: ``` Меня зовут Джульетта Мао ``` Раскрытая реплика позволяет использовать многие классические инструменты NLP для своей обработки, включая регулярные выражения, классификаторы интентов и т.д. Подробнее о том, какие ситуации и как обрабатываются моделью, смотрите в [конце страницы](#обрабатываемые-ситуации) и в [этом документе](https://huggingface.co/inkoziev/rugpt_interpreter/blob/main/%D0%92%D0%BE%D1%81%D1%81%D1%82%D0%B0%D0%BD%D0%BE%D0%B2%D0%BB%D0%B5%D0%BD%D0%B8%D0%B5%20%D0%BF%D0%BE%D0%BB%D0%BD%D1%8B%D1%85%20%D1%80%D0%B5%D0%BF%D0%BB%D0%B8%D0%BA%20%D0%B2%20%D0%B4%D0%B8%D0%B0%D0%BB%D0%BE%D0%B3%D0%B5.pdf). ## Пример использования Данная модель работает в прототипе [диалоговой системы](https://github.com/Koziev/chatbot). Она не требует для работы никакой "обвязки", пре- или постпроцессинга, помимо стандартных для моделей семейства GPT, поэтому использовать ее очень просто: ``` import torch from transformers import AutoTokenizer, AutoModelForCausalLM device = "cuda" if torch.cuda.is_available() else "cpu" model_name = "inkoziev/rugpt_interpreter" tokenizer = AutoTokenizer.from_pretrained(model_name) tokenizer.add_special_tokens({'bos_token': '<s>', 'eos_token': '</s>', 'pad_token': '<pad>'}) model = AutoModelForCausalLM.from_pretrained(model_name) model.to(device) model.eval() # На вход модели подаем последние 2-3 реплики диалога. Каждая реплика на отдельной строке, начинается с символа "-" # В конце добавляем символ "#" input_text = """<s>- Как тебя зовут? - Джульетта Мао #""" #input_text = """<s>- Что Предтечи забрали у Предшественников? #- Они узурпировали у них Мантию — защиту всего живого в галактике #""" encoded_prompt = tokenizer.encode(input_text, add_special_tokens=False, return_tensors="pt").to(device) output_sequences = model.generate(input_ids=encoded_prompt, max_length=100, num_return_sequences=1, pad_token_id=tokenizer.pad_token_id) text = tokenizer.decode(output_sequences[0].tolist(), clean_up_tokenization_spaces=True)[len(input_text)+1:] text = text[: text.find('</s>')] print(text) ``` ## Формат входных данных На вход модели подается результат токенизации для текста, составленного из 2 или 3 последних реплик диалога. Первым токеном должен быть ```<s>```. Каждая реплика должна начинаться префиксом "- ". Реплики разделяются символом перевода строки. К последней реплике, которая будет раскрываться, добавляется подстрока " #". ``` <s>- Как тебя зовут? - Джульетта Мао # ``` ## Обрабатываемые ситуации Модель разрабатывается с прицелом на использование в [чатботе](https://github.com/Koziev/chatbot). Она поддерживает некоторые типичные ситуации в читчате, которые перечислены далее. В примерах после символа ⇒ идет эталонная раскрытая реплика, которую должна сгенерировать модель. [Эллипсисы](https://ru.wikipedia.org/wiki/%D0%AD%D0%BB%D0%BB%D0%B8%D0%BF%D1%81%D0%B8%D1%81): ``` - Как же тебя зовут, а? - Меня – Стас, а тебя? ⇒ Меня зовут Стас. Как тебя зовут? ``` В редких случаях и главное слово в словосочетании может опускаться, модель попытается его восстановить: ``` - Мама, купи мне собаку. - А ты будешь за ней ухаживать? - А ты мне здоровую купи. ⇒ купи мне здоровую собаку ``` [Анафора](https://ru.wikipedia.org/wiki/%D0%90%D0%BD%D0%B0%D1%84%D0%BE%D1%80%D0%B0_(%D0%BB%D0%B8%D0%BD%D0%B3%D0%B2%D0%B8%D1%81%D1%82%D0%B8%D0%BA%D0%B0)): ``` - Ты собак любишь? - Не люблю я их ⇒ я не люблю собак ``` Иногда для раскрытия полной реплики требуется привлечение здравого смысла, модель для этого будет опираться на статистику претрейна: ``` - Мне на голову упала коробка. - А что в ней было? ⇒ что было в коробке|голове? ``` [Гэппинг](https://ru.wikipedia.org/wiki/%D0%AD%D0%BB%D0%BB%D0%B8%D0%BF%D1%81%D0%B8%D1%81#%D0%93%D1%8D%D0%BF%D0%BF%D0%B8%D0%BD%D0%B3_(en:Gapping)): ``` - Ты кошек любишь? - Их – нет ⇒ я не люблю кошек ``` Сложный гэппинг: ``` - В 25 лет вы получаете пенсию? - Не я - отец. ⇒ Я не получаю пенсию. Отец получает пенсию ``` Восстановление необязательного местоименного подлежащего (см. [pro drop](https://en.wikipedia.org/wiki/Pro-drop_language)): ``` - Согласна? - Да ⇒ я согласна ``` Модель пытается "читать между строк" и восстанавливать подразумеваемые части реплики: ``` - Ты разве ещё не ел? - Тебя ждал ⇒ я еще не ел. я ждал тебя. ``` Отрицания в диалоге: ``` - Я не прав? - Нет. (Да.) ⇒ ты не прав ``` Интерпретация не сводится к копированию слов из контекста, иногда модель должна добавить ассоциируемые с ситуацией слова: ``` - Как прошли выходные? - В Простоквашино ездила... ⇒ я на выходных ездила в Простоквашино ``` Все вышесказанное может быть в разных сочетаниях одновременно: ``` - Где твой кот? - Жена к ветеринару повезла. ⇒ жена повезла моего кота к ветеринару - Заболел? ⇒ твой кот заболел? ``` Сложные предложения: ``` - Я сварила суп, иди ешь. - Из чего? ⇒ из чего ты сварила суп? ``` Замена подлежащего производится, если это улучшает понимание реплики: ``` - Как себя чувствует твой попугай? - Бедняга умер... ⇒ мой попугай умер ``` Иногда от реплики остается только наречие, модель будет восстанавливать все остальное: ``` - Девушка, а Вы животных любите? - Очень! ⇒ я очень люблю животных ``` Форма сказуемого иногда может меняться из соображений согласованности: ``` - Рабинович, как думаете, что будет делать правительство, если завтра население разом бросит курить? - Таки, поднимут акцизы на алкоголь... ⇒ правительно поднимет акцизы на алкоголь, если завтра население разом бросит курить ``` Во всех случаях модель не выдает никакой информации, откуда она взяла подстановку для замены или заполнения в выходном тексте. На выходе получается просто текст реплики в том виде, как ее мог бы сказать человек, безо всяких дополнительных отсылок и маркеров: ``` - У тебя брат есть? - Да, есть - Где он работает? ⇒ Где работает твой брат? ``` В данном примере модель никак не сообщит нам, откуда она взяла подстановку “твой брат” для местоимения “он”. Это сильно упрощает ручную разметку обучающего корпуса и не особо мешает диалоговой системе. Во многих случаях модель приводит порядок слов к более-менее каноническому. Точнее говоря, она старается выдать текст с таким порядком слов, который обычно используют носители языка в данном контексте диалога. Если русскоговорящие предпочитают OVS вместо формального SVO, то модель будет выдавать именно OVS: ``` - У тебя штрафы были? - Нет, их никогда не было ⇒ у меня никогда не было штрафов ``` Модель обычно вставляет личные местоимения, даже если форма глагола позволяет обойтись без них: ``` - Жару любишь? - Ненавижу ее ⇒ я ненавижу жару ``` Сложносочиненные ответы разбиваются на отдельные клаузы, чтобы downstream pipeline мог обработать их последовательно: ``` - Тебя как зовут? - Кортана, а тебя как? ⇒ Меня зовут Кортана. Как тебя зовут? ``` В качестве контекста можно подавать последние 2 или 3 реплики. Более длинные отношения весьма редки, чтобы ради них усложнять датасет. Кроме того, во многих случаях достаточно применить модель рекурсивно - подать вместо исходных реплик диалога результат их раскрытия моделью: ``` - Где живешь? - В Шанхае ⇒ я живу в Шанхае - Давно? ⇒ ты давно живешь в Шанхае? - Два года уже ⇒ я уже два года живу в Шанхае - Как там погода? ⇒ как там погода в Шанхае? ``` Последнее, что хочется отметить: модель обучена **только** на диалоговых данных с короткими репликами (читчат). Она практически не способна раскрывать анафоры в художественных текстах, хотя это не ограничение модели, а особенность обучающего датасета. ### Пример работы в чате Первый столбец содержит реплики диалога [модели читчата](https://huggingface.co/inkoziev/rugpt_chitchat) с собой, второй столбец - результат раскрытия реплик моделью интерпретатора: ``` - Добрый вечер, бро! | - Чё, будем общаться? | Мы будем общаться? - Ага, а как же | Мы будем общаться - О чем хочешь говорить? | О чем ты хочешь говорить? - Давай о чем-нибудь хорошем | Я хочу говорить о чем-нибудь хорошем - Мне нравится обсуждать компьютерные игры | Мне нравится обсуждать компьютерные игры - О, компьютерные игры меня тоже интересуют | Меня тоже интересуют компьютерные игры - Ты играл в Minecraft? | Ты играл в Minecraft? - Неа, но хотел бы | Я не играл в игру Minecraft. Я хочу поиграть в игру Minecraft. ``` ### Датасет Обучающие данные без аугментации: [inkoziev/incomplete_utterance_restoration](https://huggingface.co/datasets/inkoziev/incomplete_utterance_restoration). ### Контакты Если у Вас есть какие-то вопросы по использованию этой модели, или предложения по ее улучшению - пишите мне mentalcomputing@gmail.com ### Citation: ``` @MISC{rugpt_interpreter, author = {Ilya Koziev}, title = {Incomplete Utterance Restoration in Russian Chit-Chat conversations}, url = {https://huggingface.co/inkoziev/rugpt_interpreter}, year = 2022 } ```

061c55590bf907c008782038f90d7695

Xiaoman/NER-CoNLL2003-V3

Xiaoman

bert

14

5

transformers

0

token-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

934

false

# NER-CoNLL2003-V3 This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on an unknown dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 7.961395091713594e-05 - train_batch_size: 1 - eval_batch_size: 8 - seed: 27 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 ### Training results ### Framework versions - Transformers 4.19.1 - Pytorch 1.11.0+cu113 - Datasets 2.2.1 - Tokenizers 0.12.1

1971fa6ba4e28350f7b3d0ba800babf8

spooncats/scottpilgrim

spooncats

null

19

5

diffusers

4

text-to-image

false

creativeml-openrail-m

null

2

0

['text-to-image', 'stable-diffusion']

false

true

738

false

### scottpilgrim Dreambooth model trained by spooncats with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Or you can run your new concept via `diffusers` [Colab Notebook for Inference](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_dreambooth_inference.ipynb) Sample pictures of this concept: ![0](https://media.discordapp.net/attachments/705948050144886795/1049833690672410634/vkAf5eCZcQAAAABJRU5ErkJggg.png)

ea8399f9163b51eede2f693b0320c81e

espnet/kan-bayashi_jsut_vits_accent_with_pause

espnet

null

27

98

espnet

2

text-to-speech

false

cc-by-4.0

['ja']

['jsut']

null

0

['espnet', 'audio', 'text-to-speech']

false

true

1,810

false

## ESPnet2 TTS pretrained model ### `kan-bayashi/jsut_vits_accent_with_pause` ♻️ Imported from https://zenodo.org/record/5414980/ This model was trained by kan-bayashi using jsut/tts1 recipe in [espnet](https://github.com/espnet/espnet/). ### Demo: How to use in ESPnet2 ```python # coming soon ``` ### Citing ESPnet ```BibTex @inproceedings{watanabe2018espnet, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson {Enrique Yalta Soplin} and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, title={{ESPnet}: End-to-End Speech Processing Toolkit}, year={2018}, booktitle={Proceedings of Interspeech}, pages={2207--2211}, doi={10.21437/Interspeech.2018-1456}, url={http://dx.doi.org/10.21437/Interspeech.2018-1456} } @inproceedings{hayashi2020espnet, title={{Espnet-TTS}: Unified, reproducible, and integratable open source end-to-end text-to-speech toolkit}, author={Hayashi, Tomoki and Yamamoto, Ryuichi and Inoue, Katsuki and Yoshimura, Takenori and Watanabe, Shinji and Toda, Tomoki and Takeda, Kazuya and Zhang, Yu and Tan, Xu}, booktitle={Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, pages={7654--7658}, year={2020}, organization={IEEE} } ``` or arXiv: ```bibtex @misc{watanabe2018espnet, title={ESPnet: End-to-End Speech Processing Toolkit}, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Enrique Yalta Soplin and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, year={2018}, eprint={1804.00015}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```

f564a94c75aab3a1c434ae83da698461

malteos/aspect-acl-scibert-scivocab-uncased

malteos

bert

5

9

transformers

1

null

true

false

mit

['sci', 'en', 'multilingual']

['acl-arc']

null

1

0

1

0

['classification', 'similarity']

false

true

1,133

false

# Aspect-based Document Similarity for Research Papers A `scibert-scivocab-uncased` model fine-tuned on the ACL Anthology corpus as in [Aspect-based Document Similarity for Research Papers](https://arxiv.org/abs/2010.06395). <img src="https://raw.githubusercontent.com/malteos/aspect-document-similarity/master/docrel.png"> See GitHub for more details: https://github.com/malteos/aspect-document-similarity ## Demo <a href="https://colab.research.google.com/github/malteos/aspect-document-similarity/blob/master/demo.ipynb"><img src="https://camo.githubusercontent.com/52feade06f2fecbf006889a904d221e6a730c194/68747470733a2f2f636f6c61622e72657365617263682e676f6f676c652e636f6d2f6173736574732f636f6c61622d62616467652e737667" alt="Google Colab"></a> You can try our trained models directly on Google Colab on all papers available on Semantic Scholar (via DOI, ArXiv ID, ACL ID, PubMed ID): <a href="https://colab.research.google.com/github/malteos/aspect-document-similarity/blob/master/demo.ipynb"><img src="https://raw.githubusercontent.com/malteos/aspect-document-similarity/master/demo.gif" alt="Click here for demo"></a>

eb0b34277c2085bf69797ea0640327c5

WillHeld/t5-base-adv-top_v2

WillHeld

mt5

25

1

transformers

0

text2text-generation

true

false

apache-2.0

['en']

['top_v2']

null

0

['generated_from_trainer']

true

2,184

false

# t5-base-adv-top_v2 This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on the top_v2 dataset. It achieves the following results on the evaluation set: - Loss: 0.0336 - Exact Match: 0.8540 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.001 - train_batch_size: 4 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 64 - total_train_batch_size: 256 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - training_steps: 3000 ### Training results | Training Loss | Epoch | Step | Validation Loss | Exact Match | |:-------------:|:-----:|:----:|:---------------:|:-----------:| | 1.4252 | 0.21 | 200 | 0.3381 | 0.1505 | | 0.4478 | 0.41 | 400 | 0.0673 | 0.3914 | | 0.38 | 0.62 | 600 | 0.0533 | 0.4060 | | 0.3603 | 0.82 | 800 | 0.0490 | 0.4132 | | 0.3539 | 1.03 | 1000 | 0.0420 | 0.4186 | | 0.3425 | 1.23 | 1200 | 0.0396 | 0.4219 | | 0.3373 | 1.44 | 1400 | 0.0384 | 0.4233 | | 0.3345 | 1.64 | 1600 | 0.0361 | 0.4247 | | 0.3334 | 1.85 | 1800 | 0.0357 | 0.4255 | | 0.33 | 2.05 | 2000 | 0.0361 | 0.4277 | | 0.3269 | 2.26 | 2200 | 0.0349 | 0.4278 | | 0.3262 | 2.46 | 2400 | 0.0345 | 0.4288 | | 0.324 | 2.67 | 2600 | 0.0342 | 0.4285 | | 0.3212 | 2.87 | 2800 | 0.0337 | 0.4295 | | 0.3257 | 3.08 | 3000 | 0.0336 | 0.4293 | ### Framework versions - Transformers 4.24.0 - Pytorch 1.13.0+cu117 - Datasets 2.7.0 - Tokenizers 0.13.2

15f9589c894fe1c51007d731abe39ca8

Shushant/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-ContaminationQAmodel_PubmedBERT

Shushant

bert

12

14

transformers

0

question-answering

true

false

mit

null

0

['generated_from_trainer']

true

1,759

false

# BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-ContaminationQAmodel_PubmedBERT This model is a fine-tuned version of [microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co/microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.7515 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | No log | 1.0 | 22 | 3.9518 | | No log | 2.0 | 44 | 3.2703 | | No log | 3.0 | 66 | 2.9308 | | No log | 4.0 | 88 | 2.7806 | | No log | 5.0 | 110 | 2.6926 | | No log | 6.0 | 132 | 2.7043 | | No log | 7.0 | 154 | 2.7113 | | No log | 8.0 | 176 | 2.7236 | | No log | 9.0 | 198 | 2.7559 | | No log | 10.0 | 220 | 2.7515 | ### Framework versions - Transformers 4.15.0 - Pytorch 1.10.0+cu111 - Datasets 1.17.0 - Tokenizers 0.10.3

28b53ca032816721c205264140962e86

anton-l/ddpm-ema-pokemon-64

anton-l

null

8

10

diffusers

1

null

false

apache-2.0

['en']

['huggan/pokemon']

null

0

[]

false

true

1,208

false

# ddpm-ema-pokemon-64 ## Model description This diffusion model is trained with the [🤗 Diffusers](https://github.com/huggingface/diffusers) library on the `huggan/pokemon` dataset. ## Intended uses & limitations #### How to use ```python # TODO: add an example code snippet for running this diffusion pipeline ``` #### Limitations and bias [TODO: provide examples of latent issues and potential remediations] ## Training data [TODO: describe the data used to train the model] ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 16 - eval_batch_size: 16 - gradient_accumulation_steps: 1 - optimizer: AdamW with betas=(0.95, 0.999), weight_decay=1e-06 and epsilon=1e-08 - lr_scheduler: cosine - lr_warmup_steps: 500 - ema_inv_gamma: 1.0 - ema_inv_gamma: 0.75 - ema_inv_gamma: 0.9999 - mixed_precision: no ### Training results 📈 [TensorBoard logs](https://huggingface.co/anton-l/ddpm-ema-pokemon-64/tensorboard?#scalars)

115a13909de5ad978ad07c23166d24f5

fathyshalab/massive_social-roberta-large-v1-1

fathyshalab

roberta

14

2

sentence-transformers

0

text-classification

true

false

apache-2.0

null

0

['setfit', 'sentence-transformers', 'text-classification']

false

true

1,456

false

# fathyshalab/massive_social-roberta-large-v1-1 This is a [SetFit model](https://github.com/huggingface/setfit) that can be used for text classification. The model has been trained using an efficient few-shot learning technique that involves: 1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning. 2. Training a classification head with features from the fine-tuned Sentence Transformer. ## Usage To use this model for inference, first install the SetFit library: ```bash python -m pip install setfit ``` You can then run inference as follows: ```python from setfit import SetFitModel # Download from Hub and run inference model = SetFitModel.from_pretrained("fathyshalab/massive_social-roberta-large-v1-1") # Run inference preds = model(["i loved the spiderman movie!", "pineapple on pizza is the worst 🤮"]) ``` ## BibTeX entry and citation info ```bibtex @article{https://doi.org/10.48550/arxiv.2209.11055, doi = {10.48550/ARXIV.2209.11055}, url = {https://arxiv.org/abs/2209.11055}, author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren}, keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences}, title = {Efficient Few-Shot Learning Without Prompts}, publisher = {arXiv}, year = {2022}, copyright = {Creative Commons Attribution 4.0 International} } ```

232620f94113960e4e057ce2cf3140e8

toloka/t5-large-for-text-aggregation

toloka

t5

7

33

transformers

3

summarization

true

false

apache-2.0

['en']

['toloka/CrowdSpeech']

null

0

['text aggregation', 'summarization']

false

true

2,939

false

# T5 Large for Text Aggregation ## Model description This is a T5 Large fine-tuned for crowdsourced text aggregation tasks. The model takes multiple performers' responses and yields a single aggregated response. This approach was introduced for the first time during [VLDB 2021 Crowd Science Challenge](https://crowdscience.ai/challenges/vldb21) and originally implemented at the second-place competitor's [GitHub](https://github.com/A1exRey/VLDB2021_workshop_t5). The [paper](http://ceur-ws.org/Vol-2932/short2.pdf) describing this model was presented at the [2nd Crowd Science Workshop](https://crowdscience.ai/conference_events/vldb21). ## How to use ```python from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, AutoConfig mname = "toloka/t5-large-for-text-aggregation" tokenizer = AutoTokenizer.from_pretrained(mname) model = AutoModelForSeq2SeqLM.from_pretrained(mname) input = "samplee text | sampl text | sample textt" input_ids = tokenizer.encode(input, return_tensors="pt") outputs = model.generate(input_ids) decoded = tokenizer.decode(outputs[0], skip_special_tokens=True) print(decoded) # sample text ``` ## Training data Pretrained weights were taken from the [original](https://huggingface.co/t5-large) T5 Large model by Google. For more details on the T5 architecture and training procedure see https://arxiv.org/abs/1910.10683 Model was fine-tuned on `train-clean`, `dev-clean` and `dev-other` parts of the [CrowdSpeech](https://huggingface.co/datasets/toloka/CrowdSpeech) dataset that was introduced in [our paper](https://openreview.net/forum?id=3_hgF1NAXU7&referrer=%5BAuthor%20Console%5D(%2Fgroup%3Fid%3DNeurIPS.cc%2F2021%2FTrack%2FDatasets_and_Benchmarks%2FRound1%2FAuthors%23your-submissions). ## Training procedure The model was fine-tuned for eight epochs directly following the HuggingFace summarization training [example](https://github.com/huggingface/transformers/tree/master/examples/pytorch/summarization). ## Eval results Dataset | Split | WER -----------|------------|---------- CrowdSpeech| test-clean | 4.99 CrowdSpeech| test-other | 10.61 ### BibTeX entry and citation info ```bibtex @inproceedings{Pletenev:21, author = {Pletenev, Sergey}, title = {{Noisy Text Sequences Aggregation as a Summarization Subtask}}, year = {2021}, booktitle = {Proceedings of the 2nd Crowd Science Workshop: Trust, Ethics, and Excellence in Crowdsourced Data Management at Scale}, pages = {15--20}, address = {Copenhagen, Denmark}, issn = {1613-0073}, url = {http://ceur-ws.org/Vol-2932/short2.pdf}, language = {english}, } ``` ```bibtex @misc{pavlichenko2021vox, title={Vox Populi, Vox DIY: Benchmark Dataset for Crowdsourced Audio Transcription}, author={Nikita Pavlichenko and Ivan Stelmakh and Dmitry Ustalov}, year={2021}, eprint={2107.01091}, archivePrefix={arXiv}, primaryClass={cs.SD} } ```

fda63336d5783fb1491e96b2ba23bf33

lenses/distilroberta-base-finetuned-assignment2

lenses

roberta

9

4

transformers

0

fill-mask

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,269

false

# distilroberta-base-finetuned-assignment2 This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.5976 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3.0 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | No log | 1.0 | 52 | 0.6602 | | No log | 2.0 | 104 | 0.5939 | | No log | 3.0 | 156 | 0.6450 | ### Framework versions - Transformers 4.19.2 - Pytorch 1.11.0+cu113 - Datasets 2.2.2 - Tokenizers 0.12.1

124f5edc29edaf43b7f68349dace781e

muhtasham/small-vanilla-target-glue-mnli-linear-probe

muhtasham

bert

10

3

transformers

0

text-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,829

false

# small-vanilla-target-glue-mnli-linear-probe This model is a fine-tuned version of [google/bert_uncased_L-4_H-512_A-8](https://huggingface.co/google/bert_uncased_L-4_H-512_A-8) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.0612 - Accuracy: 0.4363 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 3e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: constant - training_steps: 5000 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 1.1093 | 0.04 | 500 | 1.0875 | 0.3914 | | 1.089 | 0.08 | 1000 | 1.0814 | 0.3988 | | 1.0811 | 0.12 | 1500 | 1.0760 | 0.4113 | | 1.0753 | 0.16 | 2000 | 1.0728 | 0.4200 | | 1.0758 | 0.2 | 2500 | 1.0702 | 0.4252 | | 1.0727 | 0.24 | 3000 | 1.0684 | 0.4269 | | 1.0707 | 0.29 | 3500 | 1.0665 | 0.4295 | | 1.0702 | 0.33 | 4000 | 1.0648 | 0.4317 | | 1.0654 | 0.37 | 4500 | 1.0627 | 0.4352 | | 1.0637 | 0.41 | 5000 | 1.0612 | 0.4363 | ### Framework versions - Transformers 4.26.0.dev0 - Pytorch 1.13.0+cu116 - Datasets 2.8.1.dev0 - Tokenizers 0.13.2

27a7d524492d6ac3b47446c9cbd68e36

gababas/rraacchhiissbb

gababas

null

16

2

diffusers

0

text-to-image

false

creativeml-openrail-m

null

1

0

['text-to-image', 'stable-diffusion']

false

true

423

false

### rraacchhiissbb Dreambooth model trained by gababas with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb) Sample pictures of this concept:

3d70e3b05a9953b3a126f13fc975bb24

hassnain/wav2vec2-base-timit-demo-colab0

hassnain

wav2vec2

12

2

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,461

false

# wav2vec2-base-timit-demo-colab0 This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.1808 - Wer: 0.7734 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 1000 - num_epochs: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 4.8077 | 7.04 | 500 | 3.1554 | 1.0 | | 2.8549 | 14.08 | 1000 | 2.0683 | 1.0846 | | 1.3297 | 21.13 | 1500 | 1.2084 | 0.7984 | | 0.6725 | 28.17 | 2000 | 1.1808 | 0.7734 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.11.0+cu113 - Datasets 1.18.3 - Tokenizers 0.10.3

cf7379692883447eea27291a398a4072

henryscheible/wnli_bert-base-uncased_81_v2

henryscheible

null

13

0

null

0

null

true

false

apache-2.0

['en']

['glue']

null

0

['generated_from_trainer']

true

1,019

false

# wnli_bert-base-uncased_81_v2 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the GLUE WNLI dataset. It achieves the following results on the evaluation set: - Loss: 0.6991 - Accuracy: 0.4507 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 32 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5.0 ### Training results ### Framework versions - Transformers 4.23.1 - Pytorch 1.12.1 - Datasets 2.6.1 - Tokenizers 0.13.1

8915421ab7f853615da2a925142810b2

Seyfelislem/wspr-sm-ar

Seyfelislem

whisper

14

4

transformers

0

automatic-speech-recognition

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,458

false

# wspr-sm-ar This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4515 - Wer: 72.6173 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - training_steps: 2000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.4569 | 0.25 | 500 | 0.8556 | 105.5427 | | 0.5478 | 0.5 | 1000 | 0.7056 | 86.3373 | | 0.2269 | 0.75 | 1500 | 0.6320 | 114.2627 | | 0.1936 | 1.12 | 2000 | 0.4515 | 72.6173 | ### Framework versions - Transformers 4.27.0.dev0 - Pytorch 1.11.0 - Datasets 2.9.1.dev0 - Tokenizers 0.12.1

15a37a182ec58ca894e2259f422b0d31

gpssohi/distilbart-qgen-3-3

gpssohi

bart

10

6

transformers

2

summarization

true

false

apache-2.0

['en']

['squad']

null

0

['question-generation', 'summarization']

false

true

2,606

false

# Introduction This model checkpoint is obtained by first fine-tuning the sshleifer/distilbart-cnn-6-6 summarization checkpoint on the SQuAD dataset. After this, the 6-6 fine-tuned model is distilled down to a 3-3 model which gives us the final checkpoint. [GitHub Link for training scripts.](https://github.com/darth-c0d3r/bart-question-generation) # Usage The input format is as follows: `[answer] <s> [passage]`. The model will predict the question that corresponds to the answer from the passage. # Plot ![Distillation Run](distill_run_21.png) # Dataset The goal of Question Generation is to generate a valid and fluent question according to a given passage and the target answer. Hence, the input to the model will be a passage context and an answer, and the output / target will be the question for the given answer. Question Generation can be used in many scenarios, such as automatic tutoring systems, improving the performance of Question Answering models and enabling chat-bots to lead a conversation. The final dataset is created by taking the union of the following Question Answering Datasets. The dataset must have the following three columns: context, question, answer. ## [SQuAD](https://rajpurkar.github.io/SQuAD-explorer/) Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowd-workers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable. We use the SQuAD 1.1 variant which does not have unanswerable questions. So, every question will have a corresponding answer and vice-versa. ### Preprocessing The first step is to remove questions which don't have answers. After that, we split the train set into Train and Eval sets and treat the dev set as the test set. ### Stats **Original Dataset** | Split | Num Docs | Num Contexts | Ques w/ Ans | Ques w/o Ans | Num Unique Ans | | ----- | -------- | ------------ | ----------- | ------------ | -------------- | | Train | 442 | 19035 | 86821 | 43498 | 86821 | | Dev | 35 | 1204 | 5928 | 5945 | 10279 | **After Preprocessing** | Split | Num Rows | Context | Answer | Question | | ----- | -------- | ---------- | ------ | -------- | | Train | 80995 | 653,120,20 | 43,3,1 | 40,10,1 | | Eval | 5826 | 445,123,67 | 28,3,1 | 29,10,3 | | Test | 10297 | 629,129,25 | 29,4,1 | 31,10,3 | The numbers in the columns indicate max, avg, min number of words.

da7a96131dde64fb762792c4586eeeb5

asapp/sew-d-base-100k

asapp

sew-d

5

6

transformers

0

feature-extraction

true

false

apache-2.0

['en']

['librispeech_asr']

null

0

['speech']

false

true

1,700

false

# SEW-D-base [SEW-D by ASAPP Research](https://github.com/asappresearch/sew) The base model pretrained on 16kHz sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz. Note that this model should be fine-tuned on a downstream task, like Automatic Speech Recognition, Speaker Identification, Intent Classification, Emotion Recognition, etc... Paper: [Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition](https://arxiv.org/abs/2109.06870) Authors: Felix Wu, Kwangyoun Kim, Jing Pan, Kyu Han, Kilian Q. Weinberger, Yoav Artzi **Abstract** This paper is a study of performance-efficiency trade-offs in pre-trained models for automatic speech recognition (ASR). We focus on wav2vec 2.0, and formalize several architecture designs that influence both the model performance and its efficiency. Putting together all our observations, we introduce SEW (Squeezed and Efficient Wav2vec), a pre-trained model architecture with significant improvements along both performance and efficiency dimensions across a variety of training setups. For example, under the 100h-960h semi-supervised setup on LibriSpeech, SEW achieves a 1.9x inference speedup compared to wav2vec 2.0, with a 13.5% relative reduction in word error rate. With a similar inference time, SEW reduces word error rate by 25-50% across different model sizes. The original model can be found under https://github.com/asappresearch/sew#model-checkpoints . # Usage See [this blog](https://huggingface.co/blog/fine-tune-wav2vec2-english) for more information on how to fine-tune the model. Note that the class `Wav2Vec2ForCTC` has to be replaced by `SEWDForCTC`.

e132b70f1e539612269dae3cd758940c

Rakib/whisper-small-bn

Rakib

whisper

32

3

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['bn']

['mozilla-foundation/common_voice_11_0']

null

0

['whisper-event', 'generated_from_trainer']

true

5,079

false

# Whisper Small Bengali This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the mozilla-foundation/common_voice_11_0 bn dataset. It achieves the following results on the evaluation set: - Loss: 0.3377 - Wer: 14.4623 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 4 - eval_batch_size: 32 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 5000 - training_steps: 60000 ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:------:|:-----:|:---------------:|:-------:| | 0.2431 | 1.92 | 1000 | 0.2604 | 33.5683 | | 0.1403 | 3.83 | 2000 | 0.1703 | 23.7591 | | 0.0799 | 5.75 | 3000 | 0.1429 | 19.5394 | | 0.0411 | 7.66 | 4000 | 0.1568 | 19.0023 | | 0.0244 | 9.58 | 5000 | 0.1684 | 18.3154 | | 0.0139 | 11.49 | 6000 | 0.1856 | 17.7401 | | 0.0085 | 13.41 | 7000 | 0.2062 | 17.5263 | | 0.0063 | 15.33 | 8000 | 0.2146 | 17.2952 | | 0.0041 | 17.24 | 9000 | 0.2202 | 16.9966 | | 0.0033 | 19.16 | 10000 | 0.2163 | 16.4749 | | 0.0027 | 21.07 | 11000 | 0.2267 | 16.5334 | | 0.0031 | 22.99 | 12000 | 0.2313 | 16.4263 | | 0.0033 | 24.9 | 13000 | 0.2289 | 16.3544 | | 0.0025 | 26.82 | 14000 | 0.2384 | 16.0087 | | 0.0023 | 28.74 | 15000 | 0.2343 | 16.1089 | | 0.0027 | 30.65 | 16000 | 0.2389 | 16.1495 | | 0.0022 | 32.57 | 17000 | 0.2461 | 15.9631 | | 0.0016 | 34.48 | 18000 | 0.2364 | 15.9040 | | 0.0015 | 36.4 | 19000 | 0.2415 | 15.7161 | | 0.0009 | 38.31 | 20000 | 0.2411 | 15.3724 | | 0.0013 | 40.23 | 21000 | 0.2425 | 15.5817 | | 0.0013 | 42.15 | 22000 | 0.2469 | 15.5112 | | 0.001 | 44.06 | 23000 | 0.2549 | 15.5474 | | 0.0015 | 45.98 | 24000 | 0.2481 | 15.3624 | | 0.0013 | 47.89 | 25000 | 0.2517 | 15.5316 | | 0.0007 | 49.81 | 26000 | 0.2559 | 15.2305 | | 0.0006 | 51.72 | 27000 | 0.2567 | 15.4066 | | 0.0008 | 53.64 | 28000 | 0.2538 | 15.2464 | | 0.0009 | 55.56 | 29000 | 0.2468 | 15.1284 | | 0.0005 | 57.47 | 30000 | 0.2660 | 15.0138 | | 0.0003 | 59.39 | 31000 | 0.2594 | 14.9384 | | 0.0004 | 61.3 | 32000 | 0.2580 | 14.8814 | | 0.0006 | 63.22 | 33000 | 0.2642 | 14.9374 | | 0.0005 | 65.13 | 34000 | 0.2650 | 15.1155 | | 0.0003 | 67.05 | 35000 | 0.2660 | 14.9939 | | 0.0004 | 68.97 | 36000 | 0.2625 | 15.1031 | | 0.0002 | 70.88 | 37000 | 0.2782 | 14.8139 | | 0.0003 | 72.8 | 38000 | 0.2647 | 15.0768 | | 0.0004 | 74.71 | 39000 | 0.2665 | 14.8680 | | 0.0004 | 76.63 | 40000 | 0.2711 | 14.7966 | | 0.0001 | 78.54 | 41000 | 0.2742 | 14.8075 | | 0.0002 | 80.46 | 42000 | 0.2703 | 14.9364 | | 0.0001 | 82.38 | 43000 | 0.2733 | 14.7604 | | 0.0003 | 84.29 | 44000 | 0.2741 | 14.8209 | | 0.0 | 86.21 | 45000 | 0.2792 | 14.6046 | | 0.0 | 88.12 | 46000 | 0.2764 | 14.7356 | | 0.0 | 90.04 | 47000 | 0.2830 | 14.6874 | | 0.0 | 91.95 | 48000 | 0.2887 | 14.5630 | | 0.0 | 93.87 | 49000 | 0.2951 | 14.5803 | | 0.0 | 95.79 | 50000 | 0.3008 | 14.5476 | | 0.0 | 97.7 | 51000 | 0.3060 | 14.5188 | | 0.0 | 99.62 | 52000 | 0.3110 | 14.5248 | | 0.0 | 101.53 | 53000 | 0.3158 | 14.4985 | | 0.0 | 103.45 | 54000 | 0.3207 | 14.4980 | | 0.0 | 105.36 | 55000 | 0.3255 | 14.5124 | | 0.0 | 107.28 | 56000 | 0.3298 | 14.4945 | | 0.0 | 109.2 | 57000 | 0.3342 | 14.4752 | | 0.0 | 111.11 | 58000 | 0.3377 | 14.4623 | | 0.0 | 113.03 | 59000 | 0.3401 | 14.4856 | | 0.0 | 114.94 | 60000 | 0.3412 | 14.4896 | ### Framework versions - Transformers 4.27.0.dev0 - Pytorch 2.0.0.dev20230117+cu117 - Datasets 2.8.1.dev0 - Tokenizers 0.13.2

bc0fe790a16e0779a96b7f9be26618c3

Digitalwitness/distilgpt2-finetuned-shakespeare

Digitalwitness

gpt2

13

4

transformers

0

text-generation

false

true

false

apache-2.0

null

0

['generated_from_keras_callback']

true

1,965

false

# Digitalwitness/distilgpt2-finetuned-shakespeare This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on an unknown dataset. It achieves the following results on the evaluation set: - Train Loss: 2.0603 - Validation Loss: 2.2069 - Epoch: 19 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01} - training_precision: float32 ### Training results | Train Loss | Validation Loss | Epoch | |:----------:|:---------------:|:-----:| | 3.4056 | 3.1490 | 0 | | 3.1359 | 2.9958 | 1 | | 2.9970 | 2.9052 | 2 | | 2.9003 | 2.8363 | 3 | | 2.8192 | 2.7759 | 4 | | 2.7524 | 2.7306 | 5 | | 2.6881 | 2.6775 | 6 | | 2.6294 | 2.6329 | 7 | | 2.5716 | 2.5949 | 8 | | 2.5213 | 2.5512 | 9 | | 2.4652 | 2.5107 | 10 | | 2.4156 | 2.4803 | 11 | | 2.3677 | 2.4329 | 12 | | 2.3163 | 2.3989 | 13 | | 2.2735 | 2.3695 | 14 | | 2.2311 | 2.3317 | 15 | | 2.1842 | 2.2924 | 16 | | 2.1386 | 2.2688 | 17 | | 2.1015 | 2.2297 | 18 | | 2.0603 | 2.2069 | 19 | ### Framework versions - Transformers 4.23.1 - TensorFlow 2.9.2 - Datasets 2.6.0 - Tokenizers 0.13.1

b67fd06d8ef3565d460e64b4146ffcc8

TencentGameMate/chinese-hubert-large

TencentGameMate

hubert

6

37,250

transformers

8

feature-extraction

true

false

mit

null

0

[]

false

true

1,169

false

Pretrained on 10k hours WenetSpeech L subset. More details in [TencentGameMate/chinese_speech_pretrain](https://github.com/TencentGameMate/chinese_speech_pretrain) This model does not have a tokenizer as it was pretrained on audio alone. In order to use this model speech recognition, a tokenizer should be created and the model should be fine-tuned on labeled text data. python package: transformers==4.16.2 ```python import torch import torch.nn.functional as F import soundfile as sf from transformers import ( Wav2Vec2FeatureExtractor, HubertModel, ) model_path="" wav_path="" feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(model_path) model = HubertModel.from_pretrained(model_path) # for pretrain: Wav2Vec2ForPreTraining # model = Wav2Vec2ForPreTraining.from_pretrained(model_path) model = model.to(device) model = model.half() model.eval() wav, sr = sf.read(wav_path) input_values = feature_extractor(wav, return_tensors="pt").input_values input_values = input_values.half() input_values = input_values.to(device) with torch.no_grad(): outputs = model(input_values) last_hidden_state = outputs.last_hidden_state ```

e12f048513ec778e452868b28fd15ec4

lmqg/mt5-small-ruquad-qg

lmqg

mt5

40

73

transformers

0

text2text-generation

true

false

cc-by-4.0

['ru']

['lmqg/qg_ruquad']

null

0

['question generation']

true

6,698

false

# Model Card of `lmqg/mt5-small-ruquad-qg` This model is fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) for question generation task on the [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) (dataset_name: default) via [`lmqg`](https://github.com/asahi417/lm-question-generation). ### Overview - **Language model:** [google/mt5-small](https://huggingface.co/google/mt5-small) - **Language:** ru - **Training data:** [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) (default) - **Online Demo:** [https://autoqg.net/](https://autoqg.net/) - **Repository:** [https://github.com/asahi417/lm-question-generation](https://github.com/asahi417/lm-question-generation) - **Paper:** [https://arxiv.org/abs/2210.03992](https://arxiv.org/abs/2210.03992) ### Usage - With [`lmqg`](https://github.com/asahi417/lm-question-generation#lmqg-language-model-for-question-generation-) ```python from lmqg import TransformersQG # initialize model model = TransformersQG(language="ru", model="lmqg/mt5-small-ruquad-qg") # model prediction questions = model.generate_q(list_context="Нелишним будет отметить, что, развивая это направление, Д. И. Менделеев, поначалу априорно выдвинув идею о температуре, при которой высота мениска будет нулевой, в мае 1860 года провёл серию опытов.", list_answer="в мае 1860 года") ``` - With `transformers` ```python from transformers import pipeline pipe = pipeline("text2text-generation", "lmqg/mt5-small-ruquad-qg") output = pipe("Нелишним будет отметить, что, развивая это направление, Д. И. Менделеев, поначалу априорно выдвинув идею о температуре, при которой высота мениска будет нулевой, <hl> в мае 1860 года <hl> провёл серию опытов.") ``` ## Evaluation - ***Metric (Question Generation)***: [raw metric file](https://huggingface.co/lmqg/mt5-small-ruquad-qg/raw/main/eval/metric.first.sentence.paragraph_answer.question.lmqg_qg_ruquad.default.json) | | Score | Type | Dataset | |:-----------|--------:|:--------|:-----------------------------------------------------------------| | BERTScore | 84.27 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | Bleu_1 | 31.03 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | Bleu_2 | 24.58 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | Bleu_3 | 19.92 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | Bleu_4 | 16.31 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | METEOR | 26.39 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | MoverScore | 62.49 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | ROUGE_L | 31.39 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | - ***Metric (Question & Answer Generation, Reference Answer)***: Each question is generated from *the gold answer*. [raw metric file](https://huggingface.co/lmqg/mt5-small-ruquad-qg/raw/main/eval/metric.first.answer.paragraph.questions_answers.lmqg_qg_ruquad.default.json) | | Score | Type | Dataset | |:--------------------------------|--------:|:--------|:-----------------------------------------------------------------| | QAAlignedF1Score (BERTScore) | 90.17 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedF1Score (MoverScore) | 68.22 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedPrecision (BERTScore) | 90.17 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedPrecision (MoverScore) | 68.23 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedRecall (BERTScore) | 90.16 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedRecall (MoverScore) | 68.21 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | - ***Metric (Question & Answer Generation, Pipeline Approach)***: Each question is generated on the answer generated by [`lmqg/mt5-small-ruquad-ae`](https://huggingface.co/lmqg/mt5-small-ruquad-ae). [raw metric file](https://huggingface.co/lmqg/mt5-small-ruquad-qg/raw/main/eval_pipeline/metric.first.answer.paragraph.questions_answers.lmqg_qg_ruquad.default.lmqg_mt5-small-ruquad-ae.json) | | Score | Type | Dataset | |:--------------------------------|--------:|:--------|:-----------------------------------------------------------------| | QAAlignedF1Score (BERTScore) | 76.96 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedF1Score (MoverScore) | 55.53 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedPrecision (BERTScore) | 73.41 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedPrecision (MoverScore) | 53.24 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedRecall (BERTScore) | 81.05 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | | QAAlignedRecall (MoverScore) | 58.25 | default | [lmqg/qg_ruquad](https://huggingface.co/datasets/lmqg/qg_ruquad) | ## Training hyperparameters The following hyperparameters were used during fine-tuning: - dataset_path: lmqg/qg_ruquad - dataset_name: default - input_types: ['paragraph_answer'] - output_types: ['question'] - prefix_types: None - model: google/mt5-small - max_length: 512 - max_length_output: 32 - epoch: 5 - batch: 64 - lr: 0.001 - fp16: False - random_seed: 1 - gradient_accumulation_steps: 1 - label_smoothing: 0.15 The full configuration can be found at [fine-tuning config file](https://huggingface.co/lmqg/mt5-small-ruquad-qg/raw/main/trainer_config.json). ## Citation ``` @inproceedings{ushio-etal-2022-generative, title = "{G}enerative {L}anguage {M}odels for {P}aragraph-{L}evel {Q}uestion {G}eneration", author = "Ushio, Asahi and Alva-Manchego, Fernando and Camacho-Collados, Jose", booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing", month = dec, year = "2022", address = "Abu Dhabi, U.A.E.", publisher = "Association for Computational Linguistics", } ```

a9638dd02dfa1c7a2896c484fa957682

vumichien/whisper-medium-jp

vumichien

whisper

22

1,754

transformers

4

automatic-speech-recognition

true

false

apache-2.0

['ja']

['mozilla-foundation/common_voice_11_0']

null

0

['whisper-event', 'generated_from_trainer']

true

1,544

false

# openai/whisper-medium This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the common_voice_11_0 dataset. It achieves the following results on the evaluation set: - Loss: 0.3029 - Wer: 9.0355 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 32 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - training_steps: 5000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:-------:| | 0.0392 | 3.03 | 1000 | 0.2023 | 10.1807 | | 0.0036 | 7.01 | 2000 | 0.2478 | 9.4409 | | 0.0013 | 10.04 | 3000 | 0.2791 | 9.1014 | | 0.0002 | 14.01 | 4000 | 0.2970 | 9.0625 | | 0.0002 | 17.04 | 5000 | 0.3029 | 9.0355 | ### Framework versions - Transformers 4.26.0.dev0 - Pytorch 1.13.0+cu117 - Datasets 2.7.1.dev0 - Tokenizers 0.13.2

2013034464fb1748af580a5745d2e7d9

BeardedJohn/bert-finetuned-ner-per-v6

BeardedJohn

bert

8

15

transformers

0

token-classification

false

true

false

apache-2.0

null

0

['generated_from_keras_callback']

true

1,507

false

# bert-finetuned-ner-per-v6 This model is a fine-tuned version of [BeardedJohn/bert-ner-wikiann](https://huggingface.co/BeardedJohn/bert-ner-wikiann) on an unknown dataset. It achieves the following results on the evaluation set: - Train Loss: 0.0155 - Validation Loss: 0.0025 - Epoch: 0 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - optimizer: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 313, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000} - training_precision: mixed_float16 ### Training results | Train Loss | Validation Loss | Epoch | |:----------:|:---------------:|:-----:| | 0.0155 | 0.0025 | 0 | ### Framework versions - Transformers 4.24.0 - TensorFlow 2.11.0 - Datasets 2.9.0 - Tokenizers 0.11.0

920fbf5807b27332ec33be8fc9ebcf7e

jonatasgrosman/exp_w2v2r_es_xls-r_gender_male-5_female-5_s263

jonatasgrosman

wav2vec2

10

4

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['es']

['mozilla-foundation/common_voice_7_0']

null

0

['automatic-speech-recognition', 'es']

false

true

476

false

# exp_w2v2r_es_xls-r_gender_male-5_female-5_s263 Fine-tuned [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) for speech recognition using the train split of [Common Voice 7.0 (es)](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0). When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool.

30dcaa3ba298ffcff518a42edf39d399

sentence-transformers/msmarco-distilbert-base-dot-prod-v3

sentence-transformers

distilbert

15

3,767

sentence-transformers

1

sentence-similarity

true

false

apache-2.0

null

0

['sentence-transformers', 'feature-extraction', 'sentence-similarity', 'transformers']

false

true

2,203

false

# sentence-transformers/msmarco-distilbert-base-dot-prod-v3 This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. ## Usage (Sentence-Transformers) Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed: ``` pip install -U sentence-transformers ``` Then you can use the model like this: ```python from sentence_transformers import SentenceTransformer sentences = ["This is an example sentence", "Each sentence is converted"] model = SentenceTransformer('sentence-transformers/msmarco-distilbert-base-dot-prod-v3') embeddings = model.encode(sentences) print(embeddings) ``` ## Evaluation Results For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/msmarco-distilbert-base-dot-prod-v3) ## Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False}) (2): Dense({'in_features': 768, 'out_features': 768, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'}) ) ``` ## Citing & Authors This model was trained by [sentence-transformers](https://www.sbert.net/). If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084): ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "http://arxiv.org/abs/1908.10084", } ```

6e35ec5678e772310315ca1ca73670e7

zhangfx7/distilbert-base-uncased-finetuned-cola

zhangfx7

distilbert

13

4

transformers

0

text-classification

true

false

apache-2.0

null

['glue']

null

0

['generated_from_trainer']

true

1,275

false

# distilbert-base-uncased-finetuned-cola This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the glue dataset. It achieves the following results on the evaluation set: - Loss: 0.4908 - Matthews Correlation: 0.4468 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 ### Training results | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation | |:-------------:|:-----:|:----:|:---------------:|:--------------------:| | 0.5214 | 1.0 | 535 | 0.4908 | 0.4468 | ### Framework versions - Transformers 4.24.0 - Pytorch 1.12.1+cu113 - Datasets 2.6.1 - Tokenizers 0.13.2

b22f6b193617578d84e04fd7e99cd024

sirhugh15/xlm-roberta-base-finetuned-panx-de-fr

sirhugh15

xlm-roberta

10

8

transformers

0

token-classification

true

false

mit

null

0

['generated_from_trainer']

true

1,321

false

# xlm-roberta-base-finetuned-panx-de-fr This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1661 - F1: 0.8557 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 24 - eval_batch_size: 24 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:----:|:---------------:|:------:| | 0.2935 | 1.0 | 715 | 0.1887 | 0.8216 | | 0.1476 | 2.0 | 1430 | 0.1625 | 0.8473 | | 0.0955 | 3.0 | 2145 | 0.1661 | 0.8557 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.12.1+cu113 - Datasets 1.16.1 - Tokenizers 0.10.3

6922da407c9857b06aaba8d71b866318

s3nh/DialoGPT-small-buzz-toy-story

s3nh

gpt2

9

8

transformers

0

conversational

true

false

openrail

['en']

null

0

[]

false

true

3,580

false

<a href="https://www.buymeacoffee.com/s3nh"><img src="https://www.buymeacoffee.com/assets/img/guidelines/download-assets-sm-1.svg" alt=""></a> <img src = 'https://images.unsplash.com/photo-1599623560574-39d485900c95?ixlib=rb-4.0.3&ixid=MnwxMjA3fDB8MHxwaG90by1wYWdlfHx8fGVufDB8fHx8&auto=format&fit=crop&w=1170&q=80'> ### Description DialogGPT is a variant of the GPT (Generative Pretrained Transformer) language model developed by OpenAI. It's a deep neural network-based language model that's trained on massive amounts of text data to generate human-like text. DialogGPT uses the transformer architecture, which is a type of neural network designed for processing sequential data such as language. During the training phase, the model is exposed to a large corpus of text and learns to predict the next word in a sequence given the previous words. In the context of dialog, DialogGPT is trained to predict the response in a conversation, given the context of the conversation. This context can include one or more turns of the conversation, along with any additional information such as the topic of the conversation or the speaker's personality. At inference time, the model takes the current context of the conversation as input and generates a response. The response is generated by sampling from the model's predicted distribution over the vocabulary. Overall, DialogGPT provides a flexible and powerful solution for generating human-like text in a conversational context, allowing for the creation of a wide range of applications such as chatbots, conversational agents, and virtual assistants ## Parameters Model was trained for 20 epochs, using params as follows. ``` per_gpu_train_batch_size: int = 2 self.per_gpu_eval_batch_size: int = 2 self.gradient_accumulation_steps: int = 1 self.learning_rate: float = 5e-5 self.weight_decay: float = 0.0 self.adam_epsilon: float = 1e-8 self.max_grad_norm: int = 1.0 self.num_train_epochs: int = 20 self.max_steps: int = -1 self.warmup_steps: int = 0 self.logging_steps: int = 1000 self.save_steps: int = 3500 self.save_total_limit = None self.eval_all_checkpoints: bool = False self.no_cuda: bool = False self.overwrite_output_dir: bool = True self.overwrite_cache: bool = True self.should_continue: bool = False self.seed: int = 42 self.local_rank: int = -1 self.fp16: bool = False self.fp16_opt_level: str = 'O1' ``` ## Usage DialoGPT small version, finetuned on Buzz Scripts from Toy Story. Simple snippet of how to infer of this model: ```python from transformers import AutoModelWithLMHead, AutoModelForCausalLM, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained('s3nh/DialoGPT-small-buzz-toy-story') model = AutoModelWithLMHead.from_pretrained('s3nh/DialoGPT-small-buzz-toy-story') for step in range(4): new_user_input_ids = tokenizer.encode(input(">> User:") + tokenizer.eos_token, return_tensors='pt') bot_input_ids = torch.cat([chat_history_ids, new_user_input_ids], dim=-1) if step > 0 else new_user_input_ids chat_history_ids = model.generate( bot_input_ids, max_length=200, pad_token_id=tokenizer.eos_token_id, no_repeat_ngram_size=3, do_sample=True, top_k=100, top_p=0.7, temperature=0.8 ) print("BuzzBot: {}".format(tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)))

2ba374022b87e7c37114fe73a13580c0

MultiBertGunjanPatrick/multiberts-seed-4-180k

MultiBertGunjanPatrick

bert

7

4

transformers

0

null

true

false

apache-2.0

['en']

['bookcorpus', 'wikipedia']

null

0

['exbert', 'multiberts', 'multiberts-seed-4']

false

true

6,483

false

# MultiBERTs Seed 4 Checkpoint 180k (uncased) Seed 4 intermediate checkpoint 180k MultiBERTs (pretrained BERT) model on English language using a masked language modeling (MLM) objective. It was introduced in [this paper](https://arxiv.org/pdf/2106.16163.pdf) and first released in [this repository](https://github.com/google-research/language/tree/master/language/multiberts). This is an intermediate checkpoint. The final checkpoint can be found at [multiberts-seed-4](https://hf.co/multberts-seed-4). This model is uncased: it does not make a difference between english and English. Disclaimer: The team releasing MultiBERTs did not write a model card for this model so this model card has been written by [gchhablani](https://hf.co/gchhablani). ## Model description MultiBERTs models are transformers model pretrained on a large corpus of English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was pretrained with two objectives: - Masked language modeling (MLM): taking a sentence, the model randomly masks 15% of the words in the input then run the entire masked sentence through the model and has to predict the masked words. This is different from traditional recurrent neural networks (RNNs) that usually see the words one after the other, or from autoregressive models like GPT which internally mask the future tokens. It allows the model to learn a bidirectional representation of the sentence. - Next sentence prediction (NSP): the models concatenates two masked sentences as inputs during pretraining. Sometimes they correspond to sentences that were next to each other in the original text, sometimes not. The model then has to predict if the two sentences were following each other or not. This way, the model learns an inner representation of the English language that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard classifier using the features produced by the MultiBERTs model as inputs. ## Intended uses & limitations You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to be fine-tuned on a downstream task. See the [model hub](https://huggingface.co/models?filter=multiberts) to look for fine-tuned versions on a task that interests you. Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at model like GPT2. ### How to use Here is how to use this model to get the features of a given text in PyTorch: ```python from transformers import BertTokenizer, BertModel tokenizer = BertTokenizer.from_pretrained('multiberts-seed-4-180k') model = BertModel.from_pretrained("multiberts-seed-4-180k") text = "Replace me by any text you'd like." encoded_input = tokenizer(text, return_tensors='pt') output = model(**encoded_input) ``` ### Limitations and bias Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. This bias will also affect all fine-tuned versions of this model. For an understanding of bias of this particular checkpoint, please try out this checkpoint with the snippet present in the [Limitation and bias section](https://huggingface.co/bert-base-uncased#limitations-and-bias) of the [bert-base-uncased](https://huggingface.co/bert-base-uncased) checkpoint. ## Training data The MultiBERTs models were pretrained on [BookCorpus](https://yknzhu.wixsite.com/mbweb), a dataset consisting of 11,038 unpublished books and [English Wikipedia](https://en.wikipedia.org/wiki/English_Wikipedia) (excluding lists, tables and headers). ## Training procedure ### Preprocessing The texts are lowercased and tokenized using WordPiece and a vocabulary size of 30,000. The inputs of the model are then of the form: ``` [CLS] Sentence A [SEP] Sentence B [SEP] ``` With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two "sentences" has a combined length of less than 512 tokens. The details of the masking procedure for each sentence are the following: - 15% of the tokens are masked. - In 80% of the cases, the masked tokens are replaced by `[MASK]`. - In 10% of the cases, the masked tokens are replaced by a random token (different) from the one they replace. - In the 10% remaining cases, the masked tokens are left as is. ### Pretraining The full model was trained on 16 Cloud TPU v2 chips for two million steps with a batch size of 256. The sequence length was set to 512 throughout. The optimizer used is Adam with a learning rate of 1e-4, \$\beta_{1} = 0.9\$ and \$\beta_{2} = 0.999\$, a weight decay of 0.01, learning rate warmup for 10,000 steps and linear decay of the learning rate after. ### BibTeX entry and citation info ```bibtex @article{DBLP:journals/corr/abs-2106-16163, author = {Thibault Sellam and Steve Yadlowsky and Jason Wei and Naomi Saphra and Alexander D'Amour and Tal Linzen and Jasmijn Bastings and Iulia Turc and Jacob Eisenstein and Dipanjan Das and Ian Tenney and Ellie Pavlick}, title = {The MultiBERTs: {BERT} Reproductions for Robustness Analysis}, journal = {CoRR}, volume = {abs/2106.16163}, year = {2021}, url = {https://arxiv.org/abs/2106.16163}, eprinttype = {arXiv}, eprint = {2106.16163}, timestamp = {Mon, 05 Jul 2021 15:15:50 +0200}, biburl = {https://dblp.org/rec/journals/corr/abs-2106-16163.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} } ``` <a href="https://huggingface.co/exbert/?model=multiberts"> <img width="300px" src="https://cdn-media.huggingface.co/exbert/button.png"> </a>

4199325fb4c9a1a806f4dddc98b1dd77

ErodeesFleurs/Amtmp

ErodeesFleurs

null

6

38

espnet

0

text-to-speech

false

cc-by-4.0

['jp']

['ErodeesFleurs']

null

0

['espnet', 'audio', 'text-to-speech']

false

true

10,312

false

## ESPnet2 TTS model ### `ErodeesFleurs/Amtmp` ### Demo: How to use in ESPnet2 Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html) if you haven't done that already. ```bash cd espnet git checkout d5b5ec7b2e77bd3e10707141818b7e6c57ac6b3f pip install -e . cd egs2/amadeus/tts1 ./run.sh --skip_data_prep false --skip_train true --download_model ErodeesFleurs/Amtmp ``` ## TTS config <details><summary>expand</summary> ``` config: conf/tuning/finetune_vits.yaml print_config: false log_level: INFO dry_run: false iterator_type: sequence output_dir: exp/tts_amadeus_vits_finetune_from_jsut_32_sentence ngpu: 1 seed: 777 num_workers: 4 num_att_plot: 3 dist_backend: nccl dist_init_method: env:// dist_world_size: null dist_rank: null local_rank: 0 dist_master_addr: null dist_master_port: null dist_launcher: null multiprocessing_distributed: false unused_parameters: true sharded_ddp: false cudnn_enabled: true cudnn_benchmark: false cudnn_deterministic: false collect_stats: false write_collected_feats: false max_epoch: 2000 patience: null val_scheduler_criterion: - valid - loss early_stopping_criterion: - valid - loss - min best_model_criterion: - - train - total_count - max keep_nbest_models: 3 nbest_averaging_interval: 0 grad_clip: -1 grad_clip_type: 2.0 grad_noise: false accum_grad: 1 no_forward_run: false resume: true train_dtype: float32 use_amp: false log_interval: 50 use_matplotlib: true use_tensorboard: true create_graph_in_tensorboard: false use_wandb: true wandb_project: amadeus wandb_id: null wandb_entity: null wandb_name: null wandb_model_log_interval: -1 detect_anomaly: false pretrain_path: null init_param: - downloads/f3698edf589206588f58f5ec837fa516/exp/tts_train_vits_raw_phn_jaconv_pyopenjtalk_accent_with_pause/train.total_count.ave_10best.pth:tts:tts ignore_init_mismatch: false freeze_param: [] num_iters_per_epoch: null batch_size: 20 valid_batch_size: null batch_bins: 5000000 valid_batch_bins: null train_shape_file: - exp/tts_stats_raw_linear_spectrogram_phn_jaconv_pyopenjtalk_accent_with_pause/train/text_shape.phn - exp/tts_stats_raw_linear_spectrogram_phn_jaconv_pyopenjtalk_accent_with_pause/train/speech_shape valid_shape_file: - exp/tts_stats_raw_linear_spectrogram_phn_jaconv_pyopenjtalk_accent_with_pause/valid/text_shape.phn - exp/tts_stats_raw_linear_spectrogram_phn_jaconv_pyopenjtalk_accent_with_pause/valid/speech_shape batch_type: numel valid_batch_type: null fold_length: - 150 - 204800 sort_in_batch: descending sort_batch: descending multiple_iterator: false chunk_length: 500 chunk_shift_ratio: 0.5 num_cache_chunks: 1024 train_data_path_and_name_and_type: - - dump/22k/raw/train/text - text - text - - dump/22k/raw/train/wav.scp - speech - sound valid_data_path_and_name_and_type: - - dump/22k/raw/dev/text - text - text - - dump/22k/raw/dev/wav.scp - speech - sound allow_variable_data_keys: false max_cache_size: 0.0 max_cache_fd: 32 valid_max_cache_size: null optim: adamw optim_conf: lr: 0.0001 betas: - 0.8 - 0.99 eps: 1.0e-09 weight_decay: 0.0 scheduler: exponentiallr scheduler_conf: gamma: 0.999875 optim2: adamw optim2_conf: lr: 0.0001 betas: - 0.8 - 0.99 eps: 1.0e-09 weight_decay: 0.0 scheduler2: exponentiallr scheduler2_conf: gamma: 0.999875 generator_first: false token_list: - <blank> - <unk> - '1' - '2' - '0' - '3' - '4' - '-1' - '5' - a - o - '-2' - i - '-3' - u - e - k - n - t - '6' - r - '-4' - s - N - m - pau - '7' - sh - d - g - w - '8' - U - '-5' - I - cl - h - y - b - '9' - j - ts - ch - '-6' - z - p - '-7' - f - ky - ry - '-8' - gy - '-9' - hy - ny - '-10' - by - my - '-11' - '-12' - '-13' - py - '-14' - '-15' - v - '10' - '-16' - '-17' - '11' - '-21' - '-20' - '12' - '-19' - '13' - '-18' - '14' - dy - '15' - ty - '-22' - '16' - '18' - '19' - '17' - <sos/eos> odim: null model_conf: {} use_preprocessor: true token_type: phn bpemodel: null non_linguistic_symbols: null cleaner: jaconv g2p: pyopenjtalk_accent_with_pause feats_extract: linear_spectrogram feats_extract_conf: n_fft: 1024 hop_length: 256 win_length: null normalize: null normalize_conf: {} tts: vits tts_conf: generator_type: vits_generator generator_params: hidden_channels: 192 spks: -1 global_channels: -1 segment_size: 32 text_encoder_attention_heads: 2 text_encoder_ffn_expand: 4 text_encoder_blocks: 6 text_encoder_positionwise_layer_type: conv1d text_encoder_positionwise_conv_kernel_size: 3 text_encoder_positional_encoding_layer_type: rel_pos text_encoder_self_attention_layer_type: rel_selfattn text_encoder_activation_type: swish text_encoder_normalize_before: true text_encoder_dropout_rate: 0.1 text_encoder_positional_dropout_rate: 0.0 text_encoder_attention_dropout_rate: 0.1 use_macaron_style_in_text_encoder: true use_conformer_conv_in_text_encoder: false text_encoder_conformer_kernel_size: -1 decoder_kernel_size: 7 decoder_channels: 512 decoder_upsample_scales: - 8 - 8 - 2 - 2 decoder_upsample_kernel_sizes: - 16 - 16 - 4 - 4 decoder_resblock_kernel_sizes: - 3 - 7 - 11 decoder_resblock_dilations: - - 1 - 3 - 5 - - 1 - 3 - 5 - - 1 - 3 - 5 use_weight_norm_in_decoder: true posterior_encoder_kernel_size: 5 posterior_encoder_layers: 16 posterior_encoder_stacks: 1 posterior_encoder_base_dilation: 1 posterior_encoder_dropout_rate: 0.0 use_weight_norm_in_posterior_encoder: true flow_flows: 4 flow_kernel_size: 5 flow_base_dilation: 1 flow_layers: 4 flow_dropout_rate: 0.0 use_weight_norm_in_flow: true use_only_mean_in_flow: true stochastic_duration_predictor_kernel_size: 3 stochastic_duration_predictor_dropout_rate: 0.5 stochastic_duration_predictor_flows: 4 stochastic_duration_predictor_dds_conv_layers: 3 vocabs: 85 aux_channels: 513 discriminator_type: hifigan_multi_scale_multi_period_discriminator discriminator_params: scales: 1 scale_downsample_pooling: AvgPool1d scale_downsample_pooling_params: kernel_size: 4 stride: 2 padding: 2 scale_discriminator_params: in_channels: 1 out_channels: 1 kernel_sizes: - 15 - 41 - 5 - 3 channels: 128 max_downsample_channels: 1024 max_groups: 16 bias: true downsample_scales: - 2 - 2 - 4 - 4 - 1 nonlinear_activation: LeakyReLU nonlinear_activation_params: negative_slope: 0.1 use_weight_norm: true use_spectral_norm: false follow_official_norm: false periods: - 2 - 3 - 5 - 7 - 11 period_discriminator_params: in_channels: 1 out_channels: 1 kernel_sizes: - 5 - 3 channels: 32 downsample_scales: - 3 - 3 - 3 - 3 - 1 max_downsample_channels: 1024 bias: true nonlinear_activation: LeakyReLU nonlinear_activation_params: negative_slope: 0.1 use_weight_norm: true use_spectral_norm: false generator_adv_loss_params: average_by_discriminators: false loss_type: mse discriminator_adv_loss_params: average_by_discriminators: false loss_type: mse feat_match_loss_params: average_by_discriminators: false average_by_layers: false include_final_outputs: true mel_loss_params: fs: 22050 n_fft: 1024 hop_length: 256 win_length: null window: hann n_mels: 80 fmin: 0 fmax: null log_base: null lambda_adv: 1.0 lambda_mel: 45.0 lambda_feat_match: 2.0 lambda_dur: 1.0 lambda_kl: 1.0 sampling_rate: 22050 cache_generator_outputs: true pitch_extract: null pitch_extract_conf: {} pitch_normalize: null pitch_normalize_conf: {} energy_extract: null energy_extract_conf: {} energy_normalize: null energy_normalize_conf: {} required: - output_dir - token_list version: '202207' distributed: false ``` </details> ### Citing ESPnet ```BibTex @inproceedings{watanabe2018espnet, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, title={{ESPnet}: End-to-End Speech Processing Toolkit}, year={2018}, booktitle={Proceedings of Interspeech}, pages={2207--2211}, doi={10.21437/Interspeech.2018-1456}, url={http://dx.doi.org/10.21437/Interspeech.2018-1456} } @inproceedings{hayashi2020espnet, title={{Espnet-TTS}: Unified, reproducible, and integratable open source end-to-end text-to-speech toolkit}, author={Hayashi, Tomoki and Yamamoto, Ryuichi and Inoue, Katsuki and Yoshimura, Takenori and Watanabe, Shinji and Toda, Tomoki and Takeda, Kazuya and Zhang, Yu and Tan, Xu}, booktitle={Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, pages={7654--7658}, year={2020}, organization={IEEE} } ``` or arXiv: ```bibtex @misc{watanabe2018espnet, title={ESPnet: End-to-End Speech Processing Toolkit}, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, year={2018}, eprint={1804.00015}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```

0b1efefc1b8b1b7bda1d8f5f20656a9f

pig4431/CR_DistilBERT_5E

pig4431

distilbert

10

4

transformers

0

text-classification

true

false

apache-2.0

null

0

['generated_from_trainer']

true

2,065

false

# CR_DistilBERT_5E This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.3663 - Accuracy: 0.9 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.6345 | 0.33 | 50 | 0.5656 | 0.66 | | 0.4704 | 0.66 | 100 | 0.3705 | 0.82 | | 0.3428 | 0.99 | 150 | 0.3186 | 0.8867 | | 0.2272 | 1.32 | 200 | 0.2871 | 0.9 | | 0.259 | 1.66 | 250 | 0.2975 | 0.8867 | | 0.2583 | 1.99 | 300 | 0.3125 | 0.8867 | | 0.1713 | 2.32 | 350 | 0.3146 | 0.8867 | | 0.181 | 2.65 | 400 | 0.3602 | 0.8867 | | 0.1868 | 2.98 | 450 | 0.3319 | 0.8933 | | 0.1521 | 3.31 | 500 | 0.3413 | 0.8867 | | 0.1153 | 3.64 | 550 | 0.3868 | 0.88 | | 0.1238 | 3.97 | 600 | 0.3686 | 0.8867 | | 0.1104 | 4.3 | 650 | 0.3674 | 0.8867 | | 0.0881 | 4.64 | 700 | 0.3750 | 0.8867 | | 0.1247 | 4.97 | 750 | 0.3663 | 0.9 | ### Framework versions - Transformers 4.24.0 - Pytorch 1.13.0 - Datasets 2.3.2 - Tokenizers 0.13.1

f72b8f718602f539eb17409c6d7b7d36

StonyBrookNLP/teabreac-t5-3b-tatqa

StonyBrookNLP

t5

10

3

transformers

0

text2text-generation

true

false

cc-by-4.0

null

0

['question-answering, multi-step-reasoning, multi-hop-reasoning']

false

true

2,624

false

# What's this? This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496). This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details. We release the following models: - **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}` - **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}` - **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}` The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`. The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`. The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**. # How to use it? Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell: ```python from transformers import AutoTokenizer, AutoModelForSeq2SeqLM from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac model_name = "StonyBrookNLP/teabreac-t5-3b-tatqa" tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization model = AutoModelForSeq2SeqLM.from_pretrained(model_name) enable_digit_tokenization(tokenizer) input_texts = [ "answer_me: Who scored the first touchdown of the game?" + "context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..." # Note: some models have slightly different qn/ctxt format. See the github repo. ] input_ids = tokenizer( input_texts, return_tensors="pt", truncation=True, max_length=800, add_special_tokens=True, padding=True, )["input_ids"] generated_ids = model.generate(input_ids, min_length=1, max_length=50) generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False) generated_predictions = [ tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions ] # => ["Chaz Schilens"] ```

7dfa3e6f52e9feec3a110b7555877fcf

waifu-research-department/CC

waifu-research-department

null

3

0

null

2

null

false

mit

null

0

[]

false

true

509

false

# Description Trainer: naotsue C.C. from Code Geass # Dataset >Training: 24 images >Regularization: 500 images # Info >Model Used: Waifu Diffusion 1.2 >Steps: 3000 >Keyword: C.C (Use this in the prompt) >Class Phrase: 1girl_green_hair_yellow_eyes_anime ![Sak](https://imgs.search.brave.com/8W7I8iqTCL13jSiFJdVTeKX7bSQCT4jnAl2oBW9z1CI/rs:fit:1200:1200:1/g:ce/aHR0cHM6Ly9oZHdh/bGxwYXBlcmltLmNv/bS93cC1jb250ZW50/L3VwbG9hZHMvMjAx/Ny8wOC8yMi8xMDc4/MTItQ29kZV9HZWFz/cy1DLkMuLXNpbXBs/ZV9iYWNrZ3JvdW5k/LmpwZw)

f41e0a57de68cdfbb43f80196e474f8e

sd-concepts-library/lucky-luke

sd-concepts-library

null

13

0

null

0

null

false

mit

null

0

[]

false

true

1,462

false

### lucky-luck on Stable Diffusion This is the `<lucky-luke>` concept taught to Stable Diffusion via Textual Inversion. You can load this concept into the [Stable Conceptualizer](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/stable_conceptualizer_inference.ipynb) notebook. You can also train your own concepts and load them into the concept libraries using [this notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_textual_inversion_training.ipynb). Here is the new concept you will be able to use as an `object`: ![<lucky-luke> 0](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/1.jpeg) ![<lucky-luke> 1](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/5.jpeg) ![<lucky-luke> 2](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/7.jpeg) ![<lucky-luke> 3](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/3.jpeg) ![<lucky-luke> 4](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/2.jpeg) ![<lucky-luke> 5](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/6.jpeg) ![<lucky-luke> 6](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/0.jpeg) ![<lucky-luke> 7](https://huggingface.co/sd-concepts-library/lucky-luck/resolve/main/concept_images/4.jpeg)

741fef04a1952fd3877c34e672b92f0f

inhee/m2m100_418M-finetuned-ko-to-en4-finetuned-ko-to-en5

inhee

m2m_100

14

3

transformers

0

text2text-generation

true

false

mit

null

0

['generated_from_trainer']

true

1,697

false

# m2m100_418M-finetuned-ko-to-en4-finetuned-ko-to-en5 This model is a fine-tuned version of [inhee/m2m100_418M-finetuned-ko-to-en4](https://huggingface.co/inhee/m2m100_418M-finetuned-ko-to-en4) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2863 - Bleu: 87.4185 - Gen Len: 9.7107 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.001 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - gradient_accumulation_steps: 256 - total_train_batch_size: 1024 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:| | No log | 1.0 | 105 | 0.3571 | 78.7464 | 9.5775 | | No log | 2.0 | 210 | 0.3410 | 81.9462 | 9.6505 | | No log | 3.0 | 315 | 0.3102 | 84.746 | 9.6732 | | No log | 4.0 | 420 | 0.2929 | 86.5137 | 9.6997 | | 0.2431 | 5.0 | 525 | 0.2863 | 87.4185 | 9.7107 | ### Framework versions - Transformers 4.18.0 - Pytorch 1.11.0+cu113 - Datasets 2.1.0 - Tokenizers 0.12.1

62e24f436a5eb89347859fbb92ccc680

thomas0104/whisper_medium_nan_tw

thomas0104

whisper

19

0

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['zh']

['mozilla-foundation/common_voice_11_0']

null

0

['whisper-event', 'generated_from_trainer']

true

1,644

false

# Whisper medium nan-tw This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the mozilla-foundation/common_voice_11_0 nan-tw dataset. It achieves the following results on the evaluation set: - Loss: 0.9100 - Wer: 42.0709 - Cer: 22.3681 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 1e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - training_steps: 5000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | Cer | |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:| | 0.0568 | 5.0 | 1000 | 0.7769 | 48.2706 | 26.0890 | | 0.0057 | 10.0 | 2000 | 0.8438 | 44.0722 | 23.9270 | | 0.0041 | 15.01 | 3000 | 0.8740 | 42.8540 | 22.9554 | | 0.0001 | 20.01 | 4000 | 0.9041 | 42.1797 | 22.5496 | | 0.0001 | 25.01 | 5000 | 0.9100 | 42.0709 | 22.3681 | ### Framework versions - Transformers 4.25.1 - Pytorch 1.13.1+cu117 - Datasets 2.8.0 - Tokenizers 0.13.2

3de250973ec9791c30823e4526f0801f

NitishKumar/distilbert-base-uncased-finetuned-squad

NitishKumar

distilbert

12

2

transformers

0

question-answering

true

false

apache-2.0

null

0

['generated_from_trainer']

true

1,278

false

# distilbert-base-uncased-finetuned-squad This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 2.9423 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:-----:|:----:|:---------------:| | No log | 1.0 | 65 | 3.3894 | | No log | 2.0 | 130 | 3.0268 | | No log | 3.0 | 195 | 2.9423 | ### Framework versions - Transformers 4.21.1 - Pytorch 1.12.1+cu113 - Datasets 2.4.0 - Tokenizers 0.12.1

ed892478db29408a6f87a093c9356a9d

wietsedv/xlm-roberta-base-ft-udpos28-ca

wietsedv

xlm-roberta

8

7

transformers

0

token-classification

true

false

apache-2.0

['ca']

['universal_dependencies']

null

0

['part-of-speech', 'token-classification']

true

567

false

# XLM-RoBERTa base Universal Dependencies v2.8 POS tagging: Catalan This model is part of our paper called: - Make the Best of Cross-lingual Transfer: Evidence from POS Tagging with over 100 Languages Check the [Space](https://huggingface.co/spaces/wietsedv/xpos) for more details. ## Usage ```python from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-ca") model = AutoModelForTokenClassification.from_pretrained("wietsedv/xlm-roberta-base-ft-udpos28-ca") ```

1b1ac9d0c82f7fddb338caa4f8489616

shibing624/prompt-t5-base-chinese

shibing624

t5

11

62

transformers

4

text2text-generation

true

false

apache-2.0

['zh']

null

0

['t5', 'pytorch', 'prompt', 'zh', 'Text2Text-Generation']

false

true

6,440

false

# Chinese Prompt(prompt-t5-base-chinese) Model 中文NLP的Prompt模型[shibing624/prompt-t5-base-chinese](https://huggingface.co/shibing624/prompt-t5-base-chinese)，One model For All nlp task(OFA) 1. 在[ClueAI/PromptCLUE-base](https://huggingface.co/ClueAI/PromptCLUE-base)预训练模型上fine-tuned 了[pCLUE中文prompt数据集](https://github.com/CLUEbenchmark/pCLUE)和[SIGHAN+Wang271K中文纠错数据集](https://github.com/shibing624/pycorrector#Dataset) 2. 模型用[textgen](https://github.com/shibing624/textgen)的`T5Model`训练，复现脚本:[training_zh_prompt_model_demo.py](https://github.com/shibing624/textgen/blob/main/examples/T5/training_zh_prompt_model_demo.py) `prompt-t5-base-chinese` evaluate public test data： The overall performance of T5 on `pCLUE_test_public.json` **test**: |model|classify_score|nli_score|generate_score|mrc_f1_score|avg_score| |:-- |:--- |:--- |:--- |:--- |:--- | |ClueAI/PromptCLUE-base|0.2417|0.0|0.1731|0.2371|0.1549| |shibing624/prompt-t5-base-chinese|0.5494|0.525|0.2751|0.2259|0.3893| ## Feature PromptCLUE：大规模多任务Prompt预训练中文开源模型。千亿中文token上大规模预训练，累计学习1.5万亿中文token，支持几十个不同类型的NLP任务，具有较好的零样本学习能力和少样本学习能力。针对理解类任务，如分类、情感分析、抽取等，可以自定义标签体系；针对生成任务，可以进行多样性的文本生成。中文上的三大统一：统一模型框架，统一任务形式，统一应用方式： - 统一模型框架：采用Text-to-Text的生成式预训练模型进行统一建模。 - 统一任务形式：Prompt统一不同的NLP任务间的差异，转化为统一的text-to-text数据形式。 - 统一应用方式：对目标任务形成拿来即用的模型，下游应用时都可转化为统一的prompt自适应方式，进行zero-shot/few-shot测试。 ![arch](promptclue.png) Fine-tuned的数据集包括： 1. 单分类tnews 2. 单分类iflytek 3. 自然语言推理ocnli 4. 语义匹配afqmc 5. 指代消解-cluewsc2020 6. 关键词识别-csl 7. 阅读理解-自由式c3 8. 阅读理解-抽取式cmrc2018 9. 阅读理解-成语填空chid 10. 中文纠错数据集-sighan+wang271k ## Usage 本项目开源在文本生成项目：[textgen](https://github.com/shibing624/textgen)，可支持T5模型，通过如下命令调用： Install package: ```shell pip install -U textgen ``` ```python from textgen import T5Model model = T5Model("t5", "shibing624/prompt-t5-base-chinese") r = model.predict(["中文改错：为了让人们遵守交通规律，警查叔叔不分昼夜在忙碌。"]) print(r) # ['为了让人们遵守交通规律,警察叔叔不分昼夜在忙碌。'] ``` ## Usage (HuggingFace Transformers) Without [textgen](https://github.com/shibing624/textgen), you can use the model like this: First, you pass your input through the transformer model, then you get the generated sentence. Install package: ``` pip install transformers ``` ```python from transformers import T5ForConditionalGeneration, T5Tokenizer tokenizer = T5Tokenizer.from_pretrained("shibing624/prompt-t5-base-chinese") model = T5ForConditionalGeneration.from_pretrained("shibing624/prompt-t5-base-chinese") def batch_generate(input_texts, max_length=64): features = tokenizer(input_texts, return_tensors='pt') outputs = model.generate(input_ids=features['input_ids'], attention_mask=features['attention_mask'], max_length=max_length) return tokenizer.batch_decode(outputs, skip_special_tokens=True) r = batch_generate(["中文改错：为了让人们遵守交通规律，警查叔叔不分昼夜在忙碌。"]) print(r) ``` output: ```shell ['为了让人们遵守交通规律,警察叔叔不分昼夜在忙碌。'] ``` 模型文件组成： ``` prompt-t5-base-chinese ├── config.json ├── model_args.json ├── pytorch_model.bin ├── special_tokens_map.json ├── tokenizer_config.json ├── spiece.model └── vocab.txt ``` ## 预测示例 #### 中文改错(correction) ```bash Input: 中文改错：为了让人们遵守交通规律，警查叔叔不分昼夜在忙碌。 Model output: 为了让人们遵守交通规律,警察叔叔不分昼夜在忙碌。 ``` #### 新闻分类(classify) ```bash Input: 分类任务：折价率过低遭抛售基金泰和跌7.15%，证券时报记者朱景锋本报讯由于折价率在大盘封基中处于最低水平，基金泰和昨日遭到投资者大举抛售，跌幅达到7.15%，远超大盘。盘面显示，基金泰和随大盘高开，之后开始震荡走低，午后开始加速下行，几乎没有像样反弹。截至收盘时，在沪深300指数仅下跌2.56%的情况下，基金泰和收盘跌幅高达7.15%，在所有封基中跌幅最大，而昨日多数封基跌幅在2%左右。选项：财经，娱乐，时政，股票答案： Model output: 财经 ``` #### 意图分类(classify) ```bash Input: 意图分类：帮我定一个周日上海浦东的房间选项：闹钟，文学，酒店，艺术，体育，健康，天气，其他答案： Model output: 酒店 ``` #### 情感分析(classify) ```bash Input: 情感分析：这个看上去还可以，但其实我不喜欢选项：积极，消极答案： Model output: 消极 ``` #### 推理(generate) ```bash Input: 请推理出上下文的关系：前提：对不起事情就是这样。假设：事情就是这样，不需要道歉。选项：中立，蕴涵，矛盾答案： Model output: 矛盾 ``` #### 阅读理解(generate) ```bash Input: 阅读文章，给出答案：段落：港汇指数，全称港元实际汇兑指数（Effective Exchange Rate Index for the Hong Kong Dollar）是由香港政府统计处编制的一项指数，以反映港元与香港主要贸易伙伴之货币的名义有效汇率加权平均数的变动情况。加权比重是按1999年至2000年平均贸易模式所制定，但政府并未有公布详细的计算公式。旧港汇指数基准日为2000年1月1日，基数为100点。由2012年1月3日起，新系列港汇指数 (包括15种货币及以2010年1月 = 100) 已取代旧港汇指数系列。港汇指数的作用，主要是用于反映香港的货品及服务的价格相对于其主要贸易伙伴的变动，并通常被视作反映香港价格竞争力的指标。问题：港汇指数的加权比重如何制定？答案： Model output: 按1999年至2000年平均贸易模式所制定 ``` #### 阅读理解-自由式(generate) ```bash Input: 阅读以下对话并回答问题。男：今天怎么这么晚才来上班啊？女：昨天工作到很晚，而且我还感冒了。男：那你回去休息吧，我帮你请假。女：谢谢你。问题：女的怎么样？选项：正在工作，感冒了，在打电话，要出差。答案： Model output: 感冒了 ``` #### 摘要(generate) ```bash Input: 为下面的文章生成摘要：北京时间9月5日12时52分，四川甘孜藏族自治州泸定县发生6.8级地震。地震发生后，领导高度重视并作出重要指示，要求把抢救生命作为首要任务，全力救援受灾群众，最大限度减少人员伤亡答案： Model output: 四川甘孜发生6.8级地震 ``` #### 通用信息抽取(generate) ```bash Input: 信息抽取：据新华社电广东省清远市清城区政府昨日对外发布信息称,日前被实名举报涉嫌勒索企业、说“分分钟可以搞垮一间厂”的清城区环保局局长陈柏,已被免去清城区区委委员问题：机构名，人名，职位答案： Model output: 机构名：新华社，清城区政府，清城区环保局，清城区区委人名：陈柏职位：局长，区委委员 ``` #### 指代消解(generate) ```bash Input: 指代消解：段落：少平跟润叶进了她二爸家的院子，润生走过来对他（代词）说：“我到宿舍找了你两回，你到哪里去了？” 问题：代词“他”指代的是？答案： Model output: 少平 ``` #### 关键词抽取(generate) ```bash Input: 抽取关键词：当地时间21日，美国联邦储备委员会宣布加息75个基点，将联邦基金利率目标区间上调到3.00%至3.25%之间，符合市场预期。这是美联储今年以来第五次加息，也是连续第三次加息，创自1981年以来的最大密集加息幅度。关键词： Model output: 美联储，利率目标区间，加息，基点 ``` #### 情感倾向(classify) ```bash 文字中包含了怎样的情感：超可爱的帅哥，爱了。。。选项：厌恶，喜欢，开心，悲伤，惊讶，生气，害怕答案： Model output: 喜欢 ``` ## 训练数据集 #### 中文Prompt数据集 - 数据：[pCLUE中文prompt数据集](https://github.com/CLUEbenchmark/pCLUE) - 相关内容 - [Huggingface](https://huggingface.co/) - [PromptCLUE-base Model](https://huggingface.co/ClueAI/PromptCLUE-base) - [textgen](https://github.com/shibing624/textgen) 数据格式： ```text {"input": "哪个类别最好的描述了这篇新闻？扣篮王拉文：精彩暴扣表演！炸\n选项：故事，文化，娱乐，体育，财经，房产，汽车，教育，科技，军事，旅游，国际，股票，农业，游戏\n答案：", "target": "电竞", "answer_choices": ["故事", "文化", "娱乐", "体育", "财经", "房产", "汽车", "教育", "科技", "军事", "旅游", "国际", "股票", "农业", "游戏"], "type": "classify"} {"input": "“现在婴儿的健康状况仍很严重”记住上面的文字,考虑:“婴儿已经完全康复了。”这是总是,绝不,或有时正确的？\n答案：", "target": "绝不", "answer_choices": ["总是", "绝不", "有时"], "type": "nli"} ``` 如果需要训练Prompt模型，请参考[https://github.com/shibing624/textgen/blob/main/examples/T5/training_zh_prompt_model_demo.py](https://github.com/shibing624/textgen/blob/main/examples/T5/training_zh_prompt_model_demo.py) 附上我的训练参数： ``` epoch=5 batch_size=50 max_length=512 # input text length max_seq_length=128 # output text length ``` V100单卡训练大概48小时。 ## Citation ```latex @software{textgen, author = {Xu Ming}, title = {textgen: Implementation of Text Generation models}, year = {2022}, url = {https://github.com/shibing624/textgen}, } ```

69aed90627e8f8a7243ada920f383d23

Namig/finetuning-sentiment-model-3000-samples

Namig

distilbert

13

9

transformers

0

text-classification

true

false

apache-2.0

null

['imdb']

null

0

['generated_from_trainer']

true

1,053

false

# finetuning-sentiment-model-3000-samples This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the imdb dataset. It achieves the following results on the evaluation set: - Loss: 0.3222 - Accuracy: 0.87 - F1: 0.8704 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 2 ### Training results ### Framework versions - Transformers 4.26.0 - Pytorch 1.13.1+cu116 - Datasets 2.8.0 - Tokenizers 0.13.2

9aef1f02c8f069b0c8b5bad219e07d3c

dbmdz/flair-historic-ner-lft

dbmdz

null

4

8

flair

1

token-classification

true

false

mit

['de']

null

0

['flair', 'token-classification', 'sequence-tagger-model']

false

true

683

false

# Towards Robust Named Entity Recognition for Historic German Based on [our paper](https://www.aclweb.org/anthology/W19-4312/) we release a new model trained on the LFT dataset. **Note:** We use BPEmbeddings instead of the combination of Wikipedia, Common Crawl and character embeddings (as used in the paper), so save space and training/inferencing time. # Results | Dataset \ Run | Run 1 | Run 2 | Run 3† | Avg. | ------------- | ----- | ----- | --------- | ------------ | Development | 76.32 | 76.13 | **76.36** | 76.27 | Test | 77.07 | 77.35 | 77.20 | 77.21 Paper reported an averaged F1-score of 77.51. † denotes that this model is selected for upload.

702bf01ee6d772a1cc7879daf8cb5cad

gauravtripathy/distilbert-base-uncased-finetuned-cola

gauravtripathy

distilbert

13

3

transformers

0

text-classification

true

false

apache-2.0

null

['glue']

null

1

0

['generated_from_trainer']

true

1,571

false

# distilbert-base-uncased-finetuned-cola This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the glue dataset. It achieves the following results on the evaluation set: - Loss: 0.7550 - Matthews Correlation: 0.5265 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation | |:-------------:|:-----:|:----:|:---------------:|:--------------------:| | 0.5296 | 1.0 | 535 | 0.5144 | 0.4215 | | 0.3504 | 2.0 | 1070 | 0.4903 | 0.5046 | | 0.2393 | 3.0 | 1605 | 0.6339 | 0.5058 | | 0.175 | 4.0 | 2140 | 0.7550 | 0.5265 | | 0.1259 | 5.0 | 2675 | 0.8688 | 0.5259 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.9.0+cu111 - Datasets 1.12.1 - Tokenizers 0.10.3

26b529f8e97d6a03251207a54b67fc6b

gokuls/mobilebert_sa_GLUE_Experiment_rte_128

gokuls

mobilebert

17

4

transformers

0

text-classification

true

false

apache-2.0

['en']

['glue']

null

0

['generated_from_trainer']

true

1,581

false

# mobilebert_sa_GLUE_Experiment_rte_128 This model is a fine-tuned version of [google/mobilebert-uncased](https://huggingface.co/google/mobilebert-uncased) on the GLUE RTE dataset. It achieves the following results on the evaluation set: - Loss: 0.6926 - Accuracy: 0.5271 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 128 - eval_batch_size: 128 - seed: 10 - distributed_type: multi-GPU - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 50 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.6935 | 1.0 | 20 | 0.6926 | 0.5271 | | 0.6934 | 2.0 | 40 | 0.6930 | 0.5271 | | 0.6931 | 3.0 | 60 | 0.6932 | 0.4982 | | 0.6932 | 4.0 | 80 | 0.6929 | 0.5343 | | 0.6929 | 5.0 | 100 | 0.6945 | 0.4729 | | 0.6921 | 6.0 | 120 | 0.6929 | 0.5199 | ### Framework versions - Transformers 4.26.0 - Pytorch 1.14.0a0+410ce96 - Datasets 2.8.0 - Tokenizers 0.13.2

d7bcc1dc11ea91ebf20015f33f5eb0bf

yanekyuk/bert-uncased-keyword-discriminator

yanekyuk

bert

10

7

transformers

2

token-classification

true

false

apache-2.0

['en']

null

0

['generated_from_trainer']

true

2,974

false

# bert-uncased-keyword-discriminator This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.1296 - Precision: 0.8439 - Recall: 0.8722 - Accuracy: 0.9727 - F1: 0.8578 - Ent/precision: 0.8723 - Ent/accuracy: 0.9077 - Ent/f1: 0.8896 - Con/precision: 0.8010 - Con/accuracy: 0.8196 - Con/f1: 0.8102 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 8 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Accuracy | F1 | Ent/precision | Ent/accuracy | Ent/f1 | Con/precision | Con/accuracy | Con/f1 | |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:--------:|:------:|:-------------:|:------------:|:------:|:-------------:|:------------:|:------:| | 0.1849 | 1.0 | 1875 | 0.1323 | 0.7039 | 0.7428 | 0.9488 | 0.7228 | 0.7522 | 0.8166 | 0.7831 | 0.6268 | 0.6332 | 0.6300 | | 0.1357 | 2.0 | 3750 | 0.1132 | 0.7581 | 0.8024 | 0.9592 | 0.7796 | 0.7948 | 0.8785 | 0.8346 | 0.6971 | 0.6895 | 0.6933 | | 0.0965 | 3.0 | 5625 | 0.1033 | 0.8086 | 0.7980 | 0.9646 | 0.8032 | 0.8410 | 0.8592 | 0.8500 | 0.7560 | 0.7071 | 0.7307 | | 0.0713 | 4.0 | 7500 | 0.1040 | 0.8181 | 0.8435 | 0.9683 | 0.8306 | 0.8526 | 0.8906 | 0.8712 | 0.7652 | 0.7736 | 0.7694 | | 0.0525 | 5.0 | 9375 | 0.1126 | 0.8150 | 0.8633 | 0.9689 | 0.8385 | 0.8495 | 0.9064 | 0.8770 | 0.7629 | 0.7993 | 0.7807 | | 0.0386 | 6.0 | 11250 | 0.1183 | 0.8374 | 0.8678 | 0.9719 | 0.8523 | 0.8709 | 0.9020 | 0.8862 | 0.7877 | 0.8170 | 0.8021 | | 0.03 | 7.0 | 13125 | 0.1237 | 0.8369 | 0.8707 | 0.9723 | 0.8535 | 0.8657 | 0.9079 | 0.8863 | 0.7934 | 0.8155 | 0.8043 | | 0.0244 | 8.0 | 15000 | 0.1296 | 0.8439 | 0.8722 | 0.9727 | 0.8578 | 0.8723 | 0.9077 | 0.8896 | 0.8010 | 0.8196 | 0.8102 | ### Framework versions - Transformers 4.19.2 - Pytorch 1.11.0+cu113 - Datasets 2.2.2 - Tokenizers 0.12.1

7f1584b40f64ce63eb5587c0c8fc0211

jonatasgrosman/exp_w2v2t_ar_no-pretraining_s6

jonatasgrosman

wav2vec2

10

2

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['ar']

['mozilla-foundation/common_voice_7_0']

null

0

['automatic-speech-recognition', 'ar']

false

true

412

false

# exp_w2v2t_ar_no-pretraining_s6 Fine-tuned randomly initialized wav2vec2 model for speech recognition using the train split of [Common Voice 7.0 (ar)](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0). When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool.

9215a27c36b3580adf94451f91f1858a

spacy/ja_core_news_sm

spacy

null

27

8

spacy

0

token-classification

false

cc-by-sa-4.0

['ja']

null

0

['spacy', 'token-classification']

false

true

2,451

false

### Details: https://spacy.io/models/ja#ja_core_news_sm Japanese pipeline optimized for CPU. Components: tok2vec, morphologizer, parser, senter, ner, attribute_ruler. | Feature | Description | | --- | --- | | **Name** | `ja_core_news_sm` | | **Version** | `3.5.0` | | **spaCy** | `>=3.5.0,<3.6.0` | | **Default Pipeline** | `tok2vec`, `morphologizer`, `parser`, `attribute_ruler`, `ner` | | **Components** | `tok2vec`, `morphologizer`, `parser`, `senter`, `attribute_ruler`, `ner` | | **Vectors** | 0 keys, 0 unique vectors (0 dimensions) | | **Sources** | [UD Japanese GSD v2.8](https://github.com/UniversalDependencies/UD_Japanese-GSD) (Omura, Mai; Miyao, Yusuke; Kanayama, Hiroshi; Matsuda, Hiroshi; Wakasa, Aya; Yamashita, Kayo; Asahara, Masayuki; Tanaka, Takaaki; Murawaki, Yugo; Matsumoto, Yuji; Mori, Shinsuke; Uematsu, Sumire; McDonald, Ryan; Nivre, Joakim; Zeman, Daniel)<br />[UD Japanese GSD v2.8 NER](https://github.com/megagonlabs/UD_Japanese-GSD) (Megagon Labs Tokyo) | | **License** | `CC BY-SA 4.0` | | **Author** | [Explosion](https://explosion.ai) | ### Label Scheme <details> <summary>View label scheme (65 labels for 3 components)</summary> | Component | Labels | | --- | --- | | **`morphologizer`** | `POS=NOUN`, `POS=ADP`, `POS=VERB`, `POS=SCONJ`, `POS=AUX`, `POS=PUNCT`, `POS=PART`, `POS=DET`, `POS=NUM`, `POS=ADV`, `POS=PRON`, `POS=ADJ`, `POS=PROPN`, `POS=CCONJ`, `POS=SYM`, `POS=NOUN\|Polarity=Neg`, `POS=AUX\|Polarity=Neg`, `POS=SPACE`, `POS=INTJ`, `POS=SCONJ\|Polarity=Neg` | | **`parser`** | `ROOT`, `acl`, `advcl`, `advmod`, `amod`, `aux`, `case`, `cc`, `ccomp`, `compound`, `cop`, `csubj`, `dep`, `det`, `dislocated`, `fixed`, `mark`, `nmod`, `nsubj`, `nummod`, `obj`, `obl`, `punct` | | **`ner`** | `CARDINAL`, `DATE`, `EVENT`, `FAC`, `GPE`, `LANGUAGE`, `LAW`, `LOC`, `MONEY`, `MOVEMENT`, `NORP`, `ORDINAL`, `ORG`, `PERCENT`, `PERSON`, `PET_NAME`, `PHONE`, `PRODUCT`, `QUANTITY`, `TIME`, `TITLE_AFFIX`, `WORK_OF_ART` | </details> ### Accuracy | Type | Score | | --- | --- | | `TOKEN_ACC` | 99.37 | | `TOKEN_P` | 97.65 | | `TOKEN_R` | 97.90 | | `TOKEN_F` | 97.77 | | `POS_ACC` | 96.09 | | `MORPH_ACC` | 0.00 | | `MORPH_MICRO_P` | 34.01 | | `MORPH_MICRO_R` | 98.04 | | `MORPH_MICRO_F` | 50.51 | | `SENTS_P` | 98.63 | | `SENTS_R` | 99.21 | | `SENTS_F` | 98.92 | | `DEP_UAS` | 91.91 | | `DEP_LAS` | 90.34 | | `TAG_ACC` | 97.12 | | `LEMMA_ACC` | 96.71 | | `ENTS_P` | 68.06 | | `ENTS_R` | 55.22 | | `ENTS_F` | 60.97 |

5729397ef4afe94bfc54eb8de70d493e

espnet/kan-bayashi_ljspeech_tts_train_transformer_raw_phn_tacotron_g2p_en_no_space_train.loss.ave

espnet

null

19

1

espnet

0

text-to-speech

false

cc-by-4.0

['en']

['ljspeech']

null

0

['espnet', 'audio', 'text-to-speech']

false

true

1,862

false

## Example ESPnet2 TTS model ### `kan-bayashi/ljspeech_tts_train_transformer_raw_phn_tacotron_g2p_en_no_space_train.loss.ave` ♻️ Imported from https://zenodo.org/record/4039194/ This model was trained by kan-bayashi using ljspeech/tts1 recipe in [espnet](https://github.com/espnet/espnet/). ### Demo: How to use in ESPnet2 ```python # coming soon ``` ### Citing ESPnet ```BibTex @inproceedings{watanabe2018espnet, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson {Enrique Yalta Soplin} and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, title={{ESPnet}: End-to-End Speech Processing Toolkit}, year={2018}, booktitle={Proceedings of Interspeech}, pages={2207--2211}, doi={10.21437/Interspeech.2018-1456}, url={http://dx.doi.org/10.21437/Interspeech.2018-1456} } @inproceedings{hayashi2020espnet, title={{Espnet-TTS}: Unified, reproducible, and integratable open source end-to-end text-to-speech toolkit}, author={Hayashi, Tomoki and Yamamoto, Ryuichi and Inoue, Katsuki and Yoshimura, Takenori and Watanabe, Shinji and Toda, Tomoki and Takeda, Kazuya and Zhang, Yu and Tan, Xu}, booktitle={Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, pages={7654--7658}, year={2020}, organization={IEEE} } ``` or arXiv: ```bibtex @misc{watanabe2018espnet, title={ESPnet: End-to-End Speech Processing Toolkit}, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Enrique Yalta Soplin and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, year={2018}, eprint={1804.00015}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```

e152a4e3f1ef3211617ee49bf49d9557

Helsinki-NLP/opus-mt-sv-bzs

Helsinki-NLP

marian

10

7

transformers

0

translation

true

false

apache-2.0

null

0

['translation']

false

true

776

false

### opus-mt-sv-bzs * source languages: sv * target languages: bzs * OPUS readme: [sv-bzs](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/sv-bzs/README.md) * dataset: opus * model: transformer-align * pre-processing: normalization + SentencePiece * download original weights: [opus-2020-01-16.zip](https://object.pouta.csc.fi/OPUS-MT-models/sv-bzs/opus-2020-01-16.zip) * test set translations: [opus-2020-01-16.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/sv-bzs/opus-2020-01-16.test.txt) * test set scores: [opus-2020-01-16.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/sv-bzs/opus-2020-01-16.eval.txt) ## Benchmarks | testset | BLEU | chr-F | |-----------------------|-------|-------| | JW300.sv.bzs | 29.4 | 0.484 |

d2445b4db9e52cee665e5dfe62427cec

Gokulapriyan/vit-base-patch16-224-finetuned-eurosat

Gokulapriyan

vit

16

17

transformers

0

image-classification

true

false

apache-2.0

null

['imagefolder']

null

0

['generated_from_trainer']

true

1,586

false

# vit-base-patch16-224-finetuned-eurosat This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset. It achieves the following results on the evaluation set: - Loss: 0.0419 - Accuracy: 0.9834 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 32 - eval_batch_size: 32 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 128 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_ratio: 0.1 - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.352 | 1.0 | 527 | 0.2383 | 0.9065 | | 0.2104 | 2.0 | 1054 | 0.1154 | 0.9562 | | 0.1764 | 3.0 | 1581 | 0.0837 | 0.9703 | | 0.1646 | 4.0 | 2108 | 0.0570 | 0.9806 | | 0.1284 | 5.0 | 2635 | 0.0419 | 0.9834 | ### Framework versions - Transformers 4.26.0 - Pytorch 1.13.1+cu116 - Datasets 2.9.0 - Tokenizers 0.13.2

bd9e4e52927f133db89bdfddb03668ee

anas-awadalla/t5-small-few-shot-k-16-finetuned-squad-seed-2

anas-awadalla

t5

15

1

transformers

0

text2text-generation

true

false

apache-2.0

null

['squad']

null

0

['generated_from_trainer']

true

963

false

# t5-small-few-shot-k-16-finetuned-squad-seed-2 This model is a fine-tuned version of [google/t5-v1_1-small](https://huggingface.co/google/t5-v1_1-small) on the squad dataset. ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 4 - eval_batch_size: 8 - seed: 2 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: constant - training_steps: 1000 ### Training results ### Framework versions - Transformers 4.20.0.dev0 - Pytorch 1.11.0+cu113 - Datasets 2.3.2 - Tokenizers 0.11.6

2ec75e3ad4ae140f6adce3ad15da5e8a

SkyR/hing-mbert-ours-run-5

SkyR

bert

10

7

transformers

0

text-classification

true

false

cc-by-4.0

null

0

['generated_from_trainer']

true

3,081

false

# hing-mbert-ours-run-5 This model is a fine-tuned version of [l3cube-pune/hing-mbert](https://huggingface.co/l3cube-pune/hing-mbert) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 3.2437 - Accuracy: 0.665 - Precision: 0.6223 - Recall: 0.5991 - F1: 0.6039 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 16 - eval_batch_size: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 20 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:| | 0.9643 | 1.0 | 100 | 0.7996 | 0.69 | 0.6596 | 0.6593 | 0.6521 | | 0.6951 | 2.0 | 200 | 1.0464 | 0.66 | 0.6597 | 0.5831 | 0.5734 | | 0.4245 | 3.0 | 300 | 0.9640 | 0.64 | 0.6025 | 0.6033 | 0.6010 | | 0.238 | 4.0 | 400 | 1.6744 | 0.68 | 0.7095 | 0.6445 | 0.6359 | | 0.1477 | 5.0 | 500 | 1.7115 | 0.665 | 0.6362 | 0.6422 | 0.6360 | | 0.1206 | 6.0 | 600 | 2.0459 | 0.635 | 0.5749 | 0.5752 | 0.5726 | | 0.0528 | 7.0 | 700 | 2.5698 | 0.66 | 0.6230 | 0.5904 | 0.5985 | | 0.0525 | 8.0 | 800 | 2.2729 | 0.625 | 0.5741 | 0.5860 | 0.5733 | | 0.0174 | 9.0 | 900 | 2.6227 | 0.635 | 0.6099 | 0.6044 | 0.6019 | | 0.0088 | 10.0 | 1000 | 2.8854 | 0.63 | 0.5699 | 0.5676 | 0.5680 | | 0.0085 | 11.0 | 1100 | 3.2173 | 0.655 | 0.6043 | 0.5771 | 0.5821 | | 0.0121 | 12.0 | 1200 | 3.1270 | 0.665 | 0.6214 | 0.5903 | 0.5971 | | 0.0141 | 13.0 | 1300 | 2.6648 | 0.655 | 0.5981 | 0.5978 | 0.5961 | | 0.0116 | 14.0 | 1400 | 3.1711 | 0.665 | 0.6192 | 0.5915 | 0.5971 | | 0.007 | 15.0 | 1500 | 3.0954 | 0.66 | 0.6156 | 0.5961 | 0.6009 | | 0.0037 | 16.0 | 1600 | 3.3065 | 0.65 | 0.6027 | 0.5791 | 0.5824 | | 0.0031 | 17.0 | 1700 | 3.1715 | 0.665 | 0.6177 | 0.5999 | 0.6048 | | 0.0021 | 18.0 | 1800 | 3.1602 | 0.665 | 0.6220 | 0.6029 | 0.6082 | | 0.0021 | 19.0 | 1900 | 3.2027 | 0.655 | 0.6096 | 0.5893 | 0.5937 | | 0.0018 | 20.0 | 2000 | 3.2437 | 0.665 | 0.6223 | 0.5991 | 0.6039 | ### Framework versions - Transformers 4.25.1 - Pytorch 1.13.0+cu116 - Tokenizers 0.13.2

b76b85b095dcaaade357c025cd2615eb

cyycyy/xlm-roberta-base-finetuned-panx-en

cyycyy

xlm-roberta

10

7

transformers

0

token-classification

true

false

mit

null

['xtreme']

null

0

['generated_from_trainer']

true

1,314

false

# xlm-roberta-base-finetuned-panx-en This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the xtreme dataset. It achieves the following results on the evaluation set: - Loss: 0.4130 - F1: 0.6851 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 5e-05 - train_batch_size: 24 - eval_batch_size: 24 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:----:|:---------------:|:------:| | 1.1435 | 1.0 | 50 | 0.5604 | 0.5493 | | 0.513 | 2.0 | 100 | 0.4557 | 0.6504 | | 0.3744 | 3.0 | 150 | 0.4130 | 0.6851 | ### Framework versions - Transformers 4.11.3 - Pytorch 1.12.1 - Datasets 1.16.1 - Tokenizers 0.10.3

5ade57eef36b599b9943283ac1b32c5a

ksabeh/bert-base-uncased-attribute-correction-mlm

ksabeh

bert

8

11

transformers

0

question-answering

false

true

false

apache-2.0

null

0

['generated_from_keras_callback']

true

1,426

false

# ksabeh/bert-base-uncased-mlm-electronics-attribute-correction This model is a fine-tuned version of [ksabeh/bert-base-uncased-mlm-electronics](https://huggingface.co/ksabeh/bert-base-uncased-mlm-electronics) on an unknown dataset. It achieves the following results on the evaluation set: - Train Loss: 0.0524 - Validation Loss: 0.0520 - Epoch: 1 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - optimizer: {'name': 'Adam', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 36848, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False} - training_precision: float32 ### Training results | Train Loss | Validation Loss | Epoch | |:----------:|:---------------:|:-----:| | 0.1459 | 0.0678 | 0 | | 0.0524 | 0.0520 | 1 | ### Framework versions - Transformers 4.18.0 - TensorFlow 2.6.4 - Datasets 2.1.0 - Tokenizers 0.12.1

d6a392d5e6be2552323bc50b82c25a9d

carlosdanielhernandezmena/stt_fo_quartznet15x5_sp_ep163_100h

carlosdanielhernandezmena

null

4

1

nemo

0

automatic-speech-recognition

true

false

cc-by-4.0

['fo']

['carlosdanielhernandezmena/ravnursson_asr']

null

0

['automatic-speech-recognition', 'speech', 'audio', 'CTC', 'pytorch', 'NeMo', 'QuartzNet', 'QuartzNet15x5', 'faroese', 'faroe islands']

true

1,825

false

# stt_fo_quartznet15x5_sp_ep163_100h **NOTE! This model was trained with the NeMo version: nemo-toolkit==1.10.0** The "stt_fo_quartznet15x5_sp_ep163_100h" is an acoustic model created with NeMo which is suitable for Automatic Speech Recognition in Faroese. It is the result of fine-tuning the model ["QuartzNet15x5Base-En.nemo"](https://catalog.ngc.nvidia.com/orgs/nvidia/models/nemospeechmodels/files) with 100 hours of Faroese data developed by the [Ravnur Project](https://maltokni.fo/en/the-ravnur-project) from the Faroe Islands and curated by Carlos Mena during 2022. Most of the data is available at public repositories such as [Clarin.is](http://hdl.handle.net/20.500.12537/276) or [Hugging Face](https://huggingface.co/datasets/carlosdanielhernandezmena/ravnursson_asr). The specific corpus used to fine-tune the model is: - [The Ravnursson Corpus: Faroese Speech and Transcripts (100h34m)](http://hdl.handle.net/20.500.12537/276) The fine-tuning process was perform during November (2022) in the servers of the [Language and Voice Laboratory](https://lvl.ru.is/) at [Reykjavík University](https://en.ru.is/) (Iceland) by Carlos Daniel Hernández Mena. ```bibtex @misc{mena2022quartznet15x5faroese, title={Acoustic Model in Faroese: stt_fo_quartznet15x5_sp_ep163_100h.}, author={Hernandez Mena, Carlos Daniel}, year={2022}, url={https://huggingface.co/carlosdanielhernandezmena/stt_fo_quartznet15x5_sp_ep163_100h}, } ``` # Acknowledgements Special thanks to Jón Guðnason, head of the Language and Voice Lab for providing computational power to make this model possible. We also want to thank to the "Language Technology Programme for Icelandic 2019-2023" which is managed and coordinated by Almannarómur, and it is funded by the Icelandic Ministry of Education, Science and Culture.

08ebadde0d6dce0c6d2deaad127458c3

cammy/bart-large-cnn-weaksup-100-NOpad-early

cammy

bart

11

1

transformers

0

text2text-generation

true

false

mit

null

0

['generated_from_trainer']

true

1,558

false

# bart-large-cnn-weaksup-100-NOpad-early This model is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.0768 - Rouge1: 28.7908 - Rouge2: 10.6989 - Rougel: 20.534 - Rougelsum: 24.1294 - Gen Len: 68.5 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 1 - eval_batch_size: 1 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 3 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:| | No log | 1.0 | 100 | 1.8905 | 31.1534 | 13.7074 | 21.6489 | 27.0709 | 64.2 | | No log | 2.0 | 200 | 2.0768 | 28.7908 | 10.6989 | 20.534 | 24.1294 | 68.5 | ### Framework versions - Transformers 4.16.2 - Pytorch 1.10.2 - Datasets 1.18.3 - Tokenizers 0.11.0

bc94d442e2707208e7d5dd14bd04c706

jonatasgrosman/exp_w2v2t_pt_wav2vec2_s515

jonatasgrosman

wav2vec2

10

5

transformers

0

automatic-speech-recognition

true

false

apache-2.0

['pt']

['mozilla-foundation/common_voice_7_0']

null

0

['automatic-speech-recognition', 'pt']

false

true

456

false

# exp_w2v2t_pt_wav2vec2_s515 Fine-tuned [facebook/wav2vec2-large-lv60](https://huggingface.co/facebook/wav2vec2-large-lv60) for speech recognition using the train split of [Common Voice 7.0 (pt)](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0). When using this model, make sure that your speech input is sampled at 16kHz. This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool.

bd86329f70e86b4885680e7aeeac0c27

Salesforce/codegen-2B-nl

Salesforce

codegen

9

1,655

transformers

0

text-generation

true

false

bsd-3-clause

null

0

[]

false

true

2,786

false

# CodeGen (CodeGen-NL 2B) ## Model description CodeGen is a family of autoregressive language models for **program synthesis** from the paper: [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) by Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong. The models are originally released in [this repository](https://github.com/salesforce/CodeGen), under 3 pre-training data variants (`NL`, `Multi`, `Mono`) and 4 model size variants (`350M`, `2B`, `6B`, `16B`). The checkpoint included in this repository is denoted as **CodeGen-NL 2B** in the paper, where "NL" means it is pre-trained on the Pile and "2B" refers to the number of trainable parameters. ## Training data This checkpoint (CodeGen-NL 2B) was pre-trained on [the Pile](https://github.com/EleutherAI/the-pile), a large-scale curated dataset created by [EleutherAI](https://www.eleuther.ai/). Parts of the dataset include code data. ## Training procedure CodeGen was trained using cross-entropy loss to maximize the likelihood of sequential inputs. The family of models are trained using multiple TPU-v4-512 by Google, leveraging data and model parallelism. See Section 2.3 of the [paper](https://arxiv.org/abs/2203.13474) for more details. ## Evaluation results We evaluate our models on two code generation benchmark: HumanEval and MTPB. Please refer to the [paper](https://arxiv.org/abs/2203.13474) for more details. ## Intended Use and Limitations As an autoregressive language model, CodeGen is capable of extracting features from given natural language and programming language texts, and calculating the likelihood of them. However, the model is intended for and best at **program synthesis**, that is, generating executable code given English prompts, where the prompts should be in the form of a comment string. The model can complete partially-generated code as well. ## How to use This model can be easily loaded using the `AutoModelForCausalLM` functionality: ```python from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Salesforce/codegen-2B-nl") model = AutoModelForCausalLM.from_pretrained("Salesforce/codegen-2B-nl") text = "def hello_world():" input_ids = tokenizer(text, return_tensors="pt").input_ids generated_ids = model.generate(input_ids, max_length=128) print(tokenizer.decode(generated_ids[0], skip_special_tokens=True)) ``` ## BibTeX entry and citation info ```bibtex @article{Nijkamp2022ACP, title={A Conversational Paradigm for Program Synthesis}, author={Nijkamp, Erik and Pang, Bo and Hayashi, Hiroaki and Tu, Lifu and Wang, Huan and Zhou, Yingbo and Savarese, Silvio and Xiong, Caiming}, journal={arXiv preprint}, year={2022} } ```

a4fd769d90a018e773a1cb9a0dd2c6ae