lbl-file11-fold0

This model is a fine-tuned version of shreyashvora/lbl-file2-fold1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.8538
• Accuracy: 0.335
• F1: 0.2912
• Precision: 0.2966
• Recall: 0.335
• Accuracy Label Label 0: 0.0
• Accuracy Label Label 1: 0.0
• Accuracy Label Label 2: 0.2167
• Accuracy Label Label 3: 0.7679
• Accuracy Label Label 4: 0.23
• Accuracy Label Label 5: 0.5692
• Accuracy Label Label 6: 0.6174
• Accuracy Label Label 7: 0.1399
• Accuracy Label Label 8: 0.4322
• Accuracy Label Label 9: 0.408

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Accuracy Label Label 0	Accuracy Label Label 1	Accuracy Label Label 2	Accuracy Label Label 3	Accuracy Label Label 4	Accuracy Label Label 5	Accuracy Label Label 6	Accuracy Label Label 7	Accuracy Label Label 8	Accuracy Label Label 9
No log	0.17	25	2.0651	0.3117	0.2477	0.2859	0.3117	0.0	0.0	0.1167	0.75	0.17	0.5615	0.8522	0.0559	0.3898	0.272
No log	0.33	50	1.9252	0.3258	0.2736	0.3040	0.3258	0.0	0.0	0.4	0.7857	0.15	0.5077	0.6	0.0490	0.4661	0.344
No log	0.5	75	1.8796	0.3342	0.2954	0.3153	0.3342	0.0	0.0	0.2083	0.7768	0.26	0.5077	0.5739	0.1329	0.5	0.424
No log	0.67	100	1.8538	0.335	0.2912	0.2966	0.335	0.0	0.0	0.2167	0.7679	0.23	0.5692	0.6174	0.1399	0.4322	0.408
No log	0.83	125	1.8400	0.3417	0.2999	0.3030	0.3417	0.0	0.0	0.3333	0.7768	0.25	0.5615	0.5739	0.1259	0.4576	0.376
No log	1.0	150	1.8357	0.3442	0.3004	0.3026	0.3442	0.0	0.0	0.3667	0.7857	0.23	0.5615	0.5826	0.1259	0.4661	0.36

Framework versions

• Transformers 4.30.2
• Pytorch 2.5.1+cpu
• Datasets 3.2.0
• Tokenizers 0.13.3