README.md

metadata

license: apache-2.0
base_model: alex-miller/ODABert
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: cva-quant-weighted-classifier
    results: []

cva-quant-weighted-classifier

This model is a fine-tuned version of alex-miller/ODABert on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1854
• Accuracy: 0.9286
• F1: 0.9231
• Precision: 0.9231
• Recall: 0.9231

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-06
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.6851	1.0	4	0.6649	0.7143	0.7333	0.6471	0.8462
0.6437	2.0	8	0.6270	0.7857	0.7857	0.7333	0.8462
0.6076	3.0	12	0.5887	0.8571	0.8462	0.8462	0.8462
0.5637	4.0	16	0.5442	0.8571	0.8462	0.8462	0.8462
0.5203	5.0	20	0.4945	0.8571	0.8462	0.8462	0.8462
0.459	6.0	24	0.4436	0.8929	0.8800	0.9167	0.8462
0.4076	7.0	28	0.3976	0.8929	0.8800	0.9167	0.8462
0.3561	8.0	32	0.3545	0.8929	0.8800	0.9167	0.8462
0.3095	9.0	36	0.3180	0.8929	0.8800	0.9167	0.8462
0.2629	10.0	40	0.2887	0.8929	0.8800	0.9167	0.8462
0.2282	11.0	44	0.2677	0.8929	0.8800	0.9167	0.8462
0.2095	12.0	48	0.2493	0.8929	0.8800	0.9167	0.8462
0.1808	13.0	52	0.2326	0.8929	0.8800	0.9167	0.8462
0.1466	14.0	56	0.2186	0.8929	0.8800	0.9167	0.8462
0.1408	15.0	60	0.2075	0.8929	0.8800	0.9167	0.8462
0.1191	16.0	64	0.1993	0.8929	0.8800	0.9167	0.8462
0.1185	17.0	68	0.1935	0.8929	0.8800	0.9167	0.8462
0.1084	18.0	72	0.1891	0.8929	0.8800	0.9167	0.8462
0.1119	19.0	76	0.1865	0.9286	0.9231	0.9231	0.9231
0.1017	20.0	80	0.1854	0.9286	0.9231	0.9231	0.9231

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1