math_question_grade_detection_v2

This model is a fine-tuned version of google-bert/bert-large-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1410
• Accuracy: 0.1591
• Precision: 0.0253
• Recall: 0.1591
• F1: 0.0437

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 850

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	0.0683	50	2.0313	0.2237	0.0856	0.2237	0.1038
No log	0.1366	100	2.2165	0.1337	0.0179	0.1337	0.0316
No log	0.2049	150	2.1366	0.1722	0.0296	0.1722	0.0506
No log	0.2732	200	2.1438	0.1722	0.0296	0.1722	0.0506
No log	0.3415	250	2.1389	0.1591	0.0253	0.1591	0.0437
No log	0.4098	300	2.1450	0.1722	0.0296	0.1722	0.0506
No log	0.4781	350	2.1456	0.1722	0.0296	0.1722	0.0506
No log	0.5464	400	2.1396	0.1722	0.0296	0.1722	0.0506
No log	0.6148	450	2.1349	0.1722	0.0296	0.1722	0.0506
2.1632	0.6831	500	2.1392	0.1591	0.0253	0.1591	0.0437
2.1632	0.7514	550	2.1391	0.1722	0.0296	0.1722	0.0506
2.1632	0.8197	600	2.1391	0.1591	0.0253	0.1591	0.0437
2.1632	0.8880	650	2.1360	0.1722	0.0296	0.1722	0.0506
2.1632	0.9563	700	2.1366	0.1722	0.0296	0.1722	0.0506
2.1632	1.0246	750	2.1388	0.1591	0.0253	0.1591	0.0437
2.1632	1.0929	800	2.1410	0.1591	0.0253	0.1591	0.0437
2.1632	1.1612	850	2.1410	0.1591	0.0253	0.1591	0.0437

Framework versions

• Transformers 4.47.0
• Pytorch 2.2.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0