QWEN_2_7B_final_task2_2.0

This model is a fine-tuned version of Qwen/Qwen2-7B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3677
• Accuracy: 0.91
• Precision: 0.9390
• Recall: 0.8775
• F1 score: 0.9072

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: 0.0001
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 score
0.5932	0.5450	200	0.3686	0.8486	0.8551	0.8405	0.8477
0.3866	1.0899	400	0.3932	0.8671	0.93	0.7949	0.8571
0.306	1.6349	600	0.3703	0.8714	0.9196	0.8148	0.8640
0.2578	2.1798	800	0.3132	0.8729	0.8541	0.9003	0.8766
0.2226	2.7248	1000	0.3252	0.8986	0.9142	0.8803	0.8970
0.154	3.2698	1200	0.3590	0.9043	0.9329	0.8718	0.9013
0.1678	3.8147	1400	0.4233	0.8943	0.9571	0.8262	0.8869
0.1221	4.3597	1600	0.3445	0.8957	0.9017	0.8889	0.8953
0.086	4.9046	1800	0.3677	0.91	0.9390	0.8775	0.9072

Framework versions

• PEFT 0.11.1
• Transformers 4.44.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1