N_bert_imdb_padding60model

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: 0.7025
• Accuracy: 0.9378

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
0.2209	1.0	1563	0.2386	0.92
0.1627	2.0	3126	0.2066	0.9349
0.0902	3.0	4689	0.3334	0.9260
0.0574	4.0	6252	0.3642	0.9349
0.0448	5.0	7815	0.3988	0.9351
0.0386	6.0	9378	0.4496	0.9329
0.014	7.0	10941	0.4423	0.9358
0.0191	8.0	12504	0.4193	0.9356
0.0183	9.0	14067	0.5018	0.9337
0.0111	10.0	15630	0.5163	0.9338
0.0125	11.0	17193	0.5138	0.9360
0.0067	12.0	18756	0.5259	0.9360
0.006	13.0	20319	0.6622	0.9314
0.0042	14.0	21882	0.6154	0.9348
0.004	15.0	23445	0.6094	0.9345
0.0031	16.0	25008	0.6450	0.9356
0.0013	17.0	26571	0.6317	0.9384
0.0014	18.0	28134	0.6713	0.9386
0.0	19.0	29697	0.6938	0.9383
0.0006	20.0	31260	0.7025	0.9378

Framework versions

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