distilbert_twitterfin_padding80model

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2464
• Accuracy: 0.8643

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6875	1.0	597	0.5594	0.7714
0.4358	2.0	1194	0.4065	0.8606
0.3212	3.0	1791	0.4008	0.8652
0.2291	4.0	2388	0.5113	0.8589
0.1828	5.0	2985	0.6224	0.8526
0.0895	6.0	3582	0.7556	0.8417
0.0787	7.0	4179	0.8308	0.8539
0.0562	8.0	4776	0.8072	0.8660
0.0622	9.0	5373	0.8961	0.8564
0.0319	10.0	5970	0.9517	0.8626
0.0301	11.0	6567	1.0274	0.8610
0.0196	12.0	7164	1.0598	0.8606
0.0224	13.0	7761	1.1347	0.8673
0.0164	14.0	8358	1.1203	0.8664
0.0113	15.0	8955	1.1585	0.8647
0.007	16.0	9552	1.2019	0.8589
0.0082	17.0	10149	1.2218	0.8656
0.0043	18.0	10746	1.2438	0.8635
0.0047	19.0	11343	1.2392	0.8656
0.0031	20.0	11940	1.2464	0.8643

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3