bert_uncased_L-2_H-768_A-12-mlm-multi-emails-hq

This model is a fine-tuned version of google/bert_uncased_L-2_H-768_A-12 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.9133
• Accuracy: 0.6452

Model description

Small BERT, uncased. 155 MB.

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 5.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.3053	0.99	141	2.1758	0.6064
2.1556	1.99	282	2.0587	0.6237
2.0616	2.99	423	1.9780	0.6355
2.0084	3.99	564	1.9317	0.6422
1.9621	4.99	705	1.9133	0.6452

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 2.0.0.dev20230129+cu118
• Datasets 2.8.0
• Tokenizers 0.13.1