PardisSzah/timely_bert_5

results

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4086
• Accuracy: 0.8531
• F1: 0.2654

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.3865	0.9999	4278	0.3995	0.8571	0.2456
0.4134	2.0	8557	0.4068	0.8540	0.2894
0.408	2.9999	12835	0.4078	0.8534	0.2717
0.3942	4.0	17114	0.4090	0.8529	0.2617
0.4345	4.9994	21390	0.4086	0.8531	0.2654

Framework versions

• Transformers 4.42.3
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1