test_trainer

This model is a fine-tuned version of bert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3803
• Accuracy: 0.842

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	125	0.6786	0.608
No log	2.0	250	0.6718	0.598
No log	3.0	375	0.6695	0.574
0.6888	4.0	500	0.6583	0.599
0.6888	5.0	625	0.6262	0.684
0.6888	6.0	750	0.5936	0.697
0.6888	7.0	875	0.5520	0.721
0.6057	8.0	1000	0.5149	0.746
0.6057	9.0	1125	0.4848	0.762
0.6057	10.0	1250	0.4558	0.779
0.6057	11.0	1375	0.4346	0.793
0.4583	12.0	1500	0.4215	0.801
0.4583	13.0	1625	0.4094	0.815
0.4583	14.0	1750	0.4027	0.816
0.4583	15.0	1875	0.3962	0.82
0.3847	16.0	2000	0.3926	0.823
0.3847	17.0	2125	0.3873	0.835
0.3847	18.0	2250	0.3857	0.833
0.3847	19.0	2375	0.3823	0.836
0.3565	20.0	2500	0.3819	0.837
0.3565	21.0	2625	0.3850	0.837
0.3565	22.0	2750	0.3806	0.839
0.3565	23.0	2875	0.3801	0.84
0.3348	24.0	3000	0.3808	0.842
0.3348	25.0	3125	0.3803	0.842

Framework versions

• PEFT 0.8.2
• Transformers 4.38.0.dev0
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.1