N_roberta_imdb_padding90model

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

• Loss: 0.4435
• Accuracy: 0.951

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
0.21	1.0	1563	0.2359	0.9291
0.1649	2.0	3126	0.1754	0.9488
0.1154	3.0	4689	0.2331	0.944
0.0712	4.0	6252	0.2467	0.9473
0.0609	5.0	7815	0.3661	0.9428
0.0473	6.0	9378	0.3834	0.9435
0.0218	7.0	10941	0.4244	0.9434
0.0205	8.0	12504	0.4267	0.9446
0.0154	9.0	14067	0.3937	0.9460
0.0172	10.0	15630	0.4532	0.9476
0.0157	11.0	17193	0.4495	0.9462
0.0125	12.0	18756	0.4728	0.9452
0.0109	13.0	20319	0.4407	0.9494
0.0083	14.0	21882	0.4388	0.9474
0.0032	15.0	23445	0.4751	0.9467
0.0039	16.0	25008	0.4764	0.9481
0.0001	17.0	26571	0.4742	0.9501
0.0027	18.0	28134	0.4530	0.9509
0.0024	19.0	29697	0.4451	0.9508
0.0033	20.0	31260	0.4435	0.951

Framework versions

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