git-base-floors

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

• Loss: 0.0322
• Wer Score: 1.8463

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Score
6.8447	4.76	50	4.1895	2.2204
2.1937	9.52	100	0.4359	3.1889
0.1385	14.29	150	0.0360	0.1426
0.0184	19.05	200	0.0307	0.6185
0.0117	23.81	250	0.0287	0.1741
0.0095	28.57	300	0.0293	1.1463
0.0081	33.33	350	0.0292	0.8463
0.0061	38.1	400	0.0312	2.0333
0.0044	42.86	450	0.0322	1.9204
0.0032	47.62	500	0.0322	1.8463

Framework versions

• Transformers 4.38.2
• Pytorch 2.0.1
• Datasets 2.18.0
• Tokenizers 0.15.2