N_bert_imdb_padding40model

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

• Loss: 0.6742
• Accuracy: 0.939

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
0.2234	1.0	1563	0.2483	0.9251
0.1543	2.0	3126	0.2148	0.9323
0.0957	3.0	4689	0.2969	0.9329
0.0674	4.0	6252	0.3085	0.9369
0.035	5.0	7815	0.3765	0.9367
0.0398	6.0	9378	0.4149	0.9368
0.0215	7.0	10941	0.4424	0.9376
0.0162	8.0	12504	0.4885	0.9352
0.0113	9.0	14067	0.4668	0.935
0.0168	10.0	15630	0.5267	0.9367
0.0077	11.0	17193	0.5049	0.9378
0.0082	12.0	18756	0.5595	0.9374
0.0055	13.0	20319	0.5650	0.9341
0.0035	14.0	21882	0.6518	0.9356
0.0017	15.0	23445	0.6662	0.9385
0.0036	16.0	25008	0.6536	0.9369
0.0	17.0	26571	0.7483	0.9354
0.0003	18.0	28134	0.7027	0.9368
0.0034	19.0	29697	0.6818	0.9384
0.0008	20.0	31260	0.6742	0.939

Framework versions

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