N_distilbert_imdb_padding20model

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

• Loss: 0.7255
• Accuracy: 0.9326

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
0.2355	1.0	1563	0.2411	0.9123
0.1693	2.0	3126	0.2567	0.926
0.1002	3.0	4689	0.3082	0.9258
0.0656	4.0	6252	0.3491	0.9269
0.0524	5.0	7815	0.4136	0.9283
0.0314	6.0	9378	0.4244	0.9277
0.0258	7.0	10941	0.4654	0.9305
0.0228	8.0	12504	0.5519	0.9232
0.0161	9.0	14067	0.6312	0.9222
0.0107	10.0	15630	0.7107	0.9210
0.0127	11.0	17193	0.6169	0.9295
0.0073	12.0	18756	0.6616	0.9286
0.0063	13.0	20319	0.6746	0.9308
0.0084	14.0	21882	0.6560	0.9308
0.0034	15.0	23445	0.6365	0.9317
0.002	16.0	25008	0.7093	0.9299
0.002	17.0	26571	0.6999	0.9316
0.0027	18.0	28134	0.7114	0.9306
0.0019	19.0	29697	0.7234	0.9322
0.0027	20.0	31260	0.7255	0.9326

Framework versions

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