Keyurjotaniya007/bert-base-cased-imdb-mlm

MLM

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

• Eval loss: 4.18

• Perplexity (PPL): 50.58

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e-5

• train_batch_size: 4

• gradient_accumulation_steps=2

• seed: 42

• weight_decay=0.01

• lr_scheduler_type: linear

• warmup_ratio=0.1

• num_epochs: 2