checkpoint-6000/README.md

metadata

base_model: sentence-transformers/all-MiniLM-L6-v2
datasets:
  - youssefkhalil320/pairs_three_scores_v5
language:
  - en
library_name: sentence-transformers
license: apache-2.0
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:80000003
  - loss:CoSENTLoss
widget:
  - source_sentence: durable pvc swim ring
    sentences:
      - flaky croissant
      - urban shoes
      - warm drinks mug
  - source_sentence: iso mak retard capsules
    sentences:
      - savory baguette
      - shea butter body cream
      - softwheeled cruiser
  - source_sentence: love sandra potty
    sentences:
      - utensil holder
      - olive pants
      - headwear
  - source_sentence: dusky hair brush
    sentences:
      - back compartment laptop
      - rubber feet platter
      - honed blade knife
  - source_sentence: nkd skn
    sentences:
      - fruit fragrances nail polish remover
      - panini salmon
      - hand drawing bag

all-MiniLM-L6-v8-pair_score

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the pairs_three_scores_v5 dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • pairs_three_scores_v5
• Language: en
• License: apache-2.0

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, '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, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    'nkd skn',
    'hand drawing bag',
    'panini salmon',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

pairs_three_scores_v5

• Dataset: pairs_three_scores_v5 at 3d8c457
• Size: 80,000,003 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 3 tokens mean: 6.06 tokens max: 12 tokens	min: 3 tokens mean: 5.71 tokens max: 13 tokens	min: 0.0 mean: 0.11 max: 1.0

• Samples:

  sentence1	sentence2	score
  vanilla hair cream	free of paraben hair mask	0.5
  nourishing shampoo	cumin lemon tea	0.0
  safe materials pacifier	facial serum	0.5

• Loss: CoSENTLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  

Evaluation Dataset

pairs_three_scores_v5

• Dataset: pairs_three_scores_v5 at 3d8c457
• Size: 20,000,001 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 3 tokens mean: 6.21 tokens max: 12 tokens	min: 3 tokens mean: 5.75 tokens max: 12 tokens	min: 0.0 mean: 0.11 max: 1.0

• Samples:

  sentence1	sentence2	score
  teddy bear toy	long lasting cat food	0.0
  eva hair treatment	fresh pineapple	0.0
  soft wave hair conditioner	hybrid seat bike	0.0

• Loss: CoSENTLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 128
• per_device_eval_batch_size: 128
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• warmup_steps: 0
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• save_safetensors: True
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• no_cuda: False
• use_cpu: False
• use_mps_device: False
• seed: 42
• data_seed: None
• jit_mode_eval: False
• use_ipex: False
• bf16: False
• fp16: True
• fp16_opt_level: O1
• half_precision_backend: auto
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: 0
• ddp_backend: None
• tpu_num_cores: None
• tpu_metrics_debug: False
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• past_index: -1
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: False
• ignore_data_skip: False
• fsdp: []
• fsdp_min_num_params: 0
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
• fsdp_transformer_layer_cls_to_wrap: None
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• skip_memory_metrics: True
• use_legacy_prediction_loop: False
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• eval_do_concat_batches: True
• fp16_backend: auto
• push_to_hub_model_id: None
• push_to_hub_organization: None
• mp_parameters:
• auto_find_batch_size: False
• full_determinism: False
• torchdynamo: None
• ray_scope: last
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• dispatch_batches: None
• split_batches: None
• include_tokens_per_second: False
• include_num_input_tokens_seen: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss
0.0002	100	10.8792
0.0003	200	10.9284
0.0005	300	10.6466
0.0006	400	10.841
0.0008	500	10.8094
0.0010	600	10.4323
0.0011	700	10.3032
0.0013	800	10.4006
0.0014	900	10.4743
0.0016	1000	10.2334
0.0018	1100	10.0135
0.0019	1200	9.7874
0.0021	1300	9.7419
0.0022	1400	9.7412
0.0024	1500	9.4585
0.0026	1600	9.5339
0.0027	1700	9.4345
0.0029	1800	9.1733
0.0030	1900	8.9952
0.0032	2000	8.9669
0.0034	2100	8.8152
0.0035	2200	8.7936
0.0037	2300	8.6771
0.0038	2400	8.4648
0.0040	2500	8.5764
0.0042	2600	8.4587
0.0043	2700	8.2966
0.0045	2800	8.2329
0.0046	2900	8.1415
0.0048	3000	8.0404
0.0050	3100	7.9698
0.0051	3200	7.9205
0.0053	3300	7.8314
0.0054	3400	7.8369
0.0056	3500	7.6403
0.0058	3600	7.5842
0.0059	3700	7.5812
0.0061	3800	7.4335
0.0062	3900	7.4917
0.0064	4000	7.3204
0.0066	4100	7.2971
0.0067	4200	7.2233
0.0069	4300	7.2081
0.0070	4400	7.1364
0.0072	4500	7.0663
0.0074	4600	6.9601
0.0075	4700	6.9546
0.0077	4800	6.9019
0.0078	4900	6.8801
0.0080	5000	6.7734
0.0082	5100	6.7648
0.0083	5200	6.7498
0.0085	5300	6.6872
0.0086	5400	6.6264
0.0088	5500	6.579
0.0090	5600	6.6001
0.0091	5700	6.5971
0.0093	5800	6.4694
0.0094	5900	6.3983
0.0096	6000	6.4477

Framework Versions

• Python: 3.8.10
• Sentence Transformers: 3.1.1
• Transformers: 4.45.2
• PyTorch: 2.4.1+cu118
• Accelerate: 1.0.1
• Datasets: 3.0.1
• Tokenizers: 0.20.3

Citation

BibTeX

Sentence Transformers

@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 = "https://arxiv.org/abs/1908.10084",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}