dream_classifier

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: 0.6082
• Accuracy: 0.8661

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
No log	1.0	17	0.6040	0.8125
No log	2.0	34	0.4455	0.8482
No log	3.0	51	0.3701	0.875
No log	4.0	68	0.4277	0.8304
No log	5.0	85	0.4477	0.8393
No log	6.0	102	0.3747	0.8929
No log	7.0	119	0.4383	0.9018
No log	8.0	136	0.5102	0.875
No log	9.0	153	0.5384	0.8661
No log	10.0	170	0.6888	0.8571
No log	11.0	187	0.5788	0.8661
No log	12.0	204	0.5933	0.8571
No log	13.0	221	0.6214	0.8661
No log	14.0	238	0.6246	0.8661
No log	15.0	255	0.6033	0.8661
No log	16.0	272	0.6066	0.8661
No log	17.0	289	0.6092	0.8661
No log	18.0	306	0.6079	0.8661
No log	19.0	323	0.6080	0.8661
No log	20.0	340	0.6082	0.8661

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu118
• Datasets 2.13.0
• Tokenizers 0.13.3