crypto_fundamental_news_text_classifier-distilbert-base-uncased

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

• Loss: 0.3716
• Accuracy: 0.9194

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0913	1.0	8	1.1070	0.2097
1.0507	2.0	16	1.0611	0.4032
0.9942	3.0	24	0.9997	0.5161
0.8923	4.0	32	0.9018	0.5968
0.7789	5.0	40	0.8149	0.6774
0.675	6.0	48	0.7557	0.7903
0.6047	7.0	56	0.6935	0.7903
0.5335	8.0	64	0.6468	0.8548
0.4758	9.0	72	0.6036	0.8871
0.43	10.0	80	0.5686	0.8871
0.3939	11.0	88	0.5312	0.9032
0.349	12.0	96	0.4888	0.9194
0.3127	13.0	104	0.4539	0.9194
0.2806	14.0	112	0.4281	0.9194
0.2624	15.0	120	0.4062	0.9194
0.2362	16.0	128	0.3953	0.9194
0.2231	17.0	136	0.3839	0.9194
0.2161	18.0	144	0.3799	0.9194
0.2023	19.0	152	0.3753	0.9194
0.1982	20.0	160	0.3716	0.9194

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 3.0.2
• Tokenizers 0.20.0