c7ed01b6a1ef5ed6a21ccd62f6018346

This model is a fine-tuned version of FacebookAI/xlm-roberta-large-finetuned-conll02-spanish on the dair-ai/emotion [split] dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
num_epochs: 50

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro
No log	0	0	1.6298	0	3.5360	0.3488	0.0862
No log	1	500	1.6232	0.0078	4.3342	0.3488	0.0862
No log	2	1000	1.5751	0.0156	5.8609	0.3488	0.0862
No log	3	1500	1.9550	0.0312	8.6197	0.0333	0.0107
No log	4	2000	1.5934	0.0625	12.4778	0.2908	0.0751
0.0882	5	2500	1.5811	0.125	18.6386	0.2908	0.0751
1.6123	6	3000	1.5741	0.25	29.2144	0.2908	0.0751
0.2615	7	3500	1.5735	0.5	49.4945	0.3488	0.0862
1.5953	8.0	4000	1.5688	1.0	88.8370	0.2908	0.0751
1.5971	9.0	4500	1.5624	1.0	87.1035	0.3488	0.0862
1.6038	10.0	5000	1.5614	1.0	87.4613	0.3488	0.0862
1.595	11.0	5500	1.5655	1.0	87.0542	0.3488	0.0862
1.5724	12.0	6000	1.5619	1.0	87.3459	0.3488	0.0862
1.5706	13.0	6500	1.5597	1.0	86.4399	0.3488	0.0862
1.5884	14.0	7000	1.5603	1.0	86.4507	0.3488	0.0862
1.585	15.0	7500	1.5641	1.0	86.4194	0.2908	0.0751
1.588	16.0	8000	1.5586	1.0	86.8251	0.3488	0.0862
1.5765	17.0	8500	1.5590	1.0	85.3048	0.3488	0.0862
1.5986	18.0	9000	1.5671	1.0	84.3959	0.2908	0.0751
1.6031	19.0	9500	1.5589	1.0	83.7903	0.3488	0.0862
1.5928	20.0	10000	1.5614	1.0	83.5218	0.3488	0.0862

Safetensors

Model size

0.6B params

Tensor type

F32

Base model

Finetuned

(21)

this model