hf-distilbert-imdb-mlm

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: 2.0927

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: 64
• eval_batch_size: 64
• seed: 420
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20.0

Training results

Training Loss	Epoch	Step	Validation Loss
2.5049	1.0	383	2.3056
2.3896	2.0	766	2.2460
2.3458	3.0	1149	2.2351
2.3097	4.0	1532	2.1917
2.2839	5.0	1915	2.1935
2.2611	6.0	2298	2.1741
2.2397	7.0	2681	2.1516
2.2234	8.0	3064	2.1464
2.2121	9.0	3447	2.1242
2.2041	10.0	3830	2.1361
2.1883	11.0	4213	2.1251
2.185	12.0	4596	2.1297
2.1712	13.0	4979	2.1062
2.1648	14.0	5362	2.1049
2.1587	15.0	5745	2.1066
2.1532	16.0	6128	2.0981
2.1472	17.0	6511	2.0926
2.1462	18.0	6894	2.0832
2.1437	19.0	7277	2.0937
2.1386	20.0	7660	2.0927

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3