sara-m98/DISTILBERT_BIRADS_ECO_MAMO_2_DESCARTADO

DISTILBERT_SARA_BIRADS_ECO_MAMO_2

This model is a fine-tuned version of distilbert/distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7336
• Accuracy: 0.9361

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
• lr_scheduler_warmup_steps: 500
• num_epochs: 32

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.8441	1.0	776	0.5188	0.8516
0.3666	2.0	1552	0.4621	0.8819
0.411	3.0	2328	0.4225	0.8877
0.2854	4.0	3104	0.4847	0.8832
0.4759	5.0	3880	0.4549	0.8948
0.1157	6.0	4656	0.4045	0.9052
0.0939	7.0	5432	0.4246	0.9110
0.086	8.0	6208	0.4076	0.9194
0.011	9.0	6984	0.4956	0.9181
0.1622	10.0	7760	0.4949	0.9155
0.0621	11.0	8536	0.4278	0.9342
0.015	12.0	9312	0.5024	0.9290
0.0308	13.0	10088	0.4937	0.9374
0.1366	14.0	10864	0.5544	0.9335
0.1376	15.0	11640	0.5920	0.9284
0.0003	16.0	12416	0.6138	0.9297
0.0008	17.0	13192	0.5690	0.9394
0.001	18.0	13968	0.5802	0.9342
0.0002	19.0	14744	0.5803	0.9368
0.0001	20.0	15520	0.6447	0.9348
0.0003	21.0	16296	0.6403	0.9374
0.0076	22.0	17072	0.6709	0.9355
0.0001	23.0	17848	0.6706	0.9368
0.0001	24.0	18624	0.6858	0.9361
0.0	25.0	19400	0.6970	0.9348
0.0	26.0	20176	0.7349	0.9316
0.0	27.0	20952	0.7660	0.9329
0.0	28.0	21728	0.7639	0.9342
0.0	29.0	22504	0.7218	0.9355
0.0001	30.0	23280	0.7302	0.9335
0.0	31.0	24056	0.7344	0.9355
0.0	32.0	24832	0.7336	0.9361

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1