scenario-KD-SCR-CDF-EN-FROM-CL-D2_data-en-cardiff_eng_only66

This model is a fine-tuned version of haryoaw/scenario-MDBT-TCR_data-cl-cardiff_cl_only on the None dataset. It achieves the following results on the evaluation set:

• Loss: 533.1400
• Accuracy: 0.3435
• F1: 0.2746

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: 32
• seed: 66
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.7391	100	630.5721	0.3347	0.2738
No log	3.4783	200	605.2021	0.3329	0.2631
No log	5.2174	300	588.4736	0.3289	0.1894
No log	6.9565	400	579.8921	0.3302	0.2716
573.1106	8.6957	500	571.2699	0.3510	0.2838
573.1106	10.4348	600	564.0858	0.3422	0.2638
573.1106	12.1739	700	560.2971	0.3364	0.2551
573.1106	13.9130	800	553.0224	0.3457	0.2748
573.1106	15.6522	900	549.0078	0.3426	0.2771
468.2501	17.3913	1000	545.7394	0.3435	0.2815
468.2501	19.1304	1100	541.9386	0.3496	0.2813
468.2501	20.8696	1200	539.8362	0.3519	0.2784
468.2501	22.6087	1300	537.6076	0.3567	0.2837
468.2501	24.3478	1400	535.7779	0.3519	0.2832
433.4717	26.0870	1500	533.8346	0.3580	0.2865
433.4717	27.8261	1600	533.4024	0.3430	0.2749
433.4717	29.5652	1700	533.1400	0.3435	0.2746

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.1+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1