distilbert-base-uncased_fold_4_binary_v1

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

• Loss: 1.5144
• F1: 0.8245

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	289	0.3756	0.8175
0.3977	2.0	578	0.3672	0.8336
0.3977	3.0	867	0.4997	0.8276
0.1972	4.0	1156	0.6597	0.8244
0.1972	5.0	1445	0.8501	0.8195
0.0824	6.0	1734	1.0074	0.8097
0.037	7.0	2023	1.1122	0.8131
0.037	8.0	2312	1.0963	0.8189
0.0182	9.0	2601	1.2511	0.8125
0.0182	10.0	2890	1.2255	0.8141
0.0121	11.0	3179	1.3120	0.8187
0.0121	12.0	3468	1.4182	0.8165
0.0079	13.0	3757	1.4142	0.8218
0.0081	14.0	4046	1.4765	0.8150
0.0081	15.0	4335	1.3510	0.8187
0.0109	16.0	4624	1.3455	0.8255
0.0109	17.0	4913	1.4157	0.8234
0.0022	18.0	5202	1.4651	0.8197
0.0022	19.0	5491	1.4388	0.8267
0.0017	20.0	5780	1.4552	0.8304
0.0005	21.0	6069	1.5357	0.8248
0.0005	22.0	6358	1.4924	0.8241
0.0009	23.0	6647	1.4865	0.8248
0.0009	24.0	6936	1.4697	0.8275
0.0013	25.0	7225	1.5144	0.8245

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1