scenario-NON-KD-SCR-COPY-CDF-EN-D2_data-en-cardiff_eng_only66

This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 6.3701
• Accuracy: 0.3404
• F1: 0.3215

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: 32
• eval_batch_size: 32
• seed: 66
• 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	1.7241	100	1.1773	0.3338	0.3116
No log	3.4483	200	2.0311	0.3461	0.3332
No log	5.1724	300	2.8249	0.3452	0.3174
No log	6.8966	400	3.6265	0.3492	0.3321
0.4477	8.6207	500	4.4107	0.3435	0.3229
0.4477	10.3448	600	5.1628	0.3413	0.2979
0.4477	12.0690	700	5.3816	0.3369	0.3166
0.4477	13.7931	800	5.4753	0.3422	0.3192
0.4477	15.5172	900	5.7827	0.3373	0.3027
0.0168	17.2414	1000	5.9046	0.3461	0.3254
0.0168	18.9655	1100	6.0785	0.3448	0.3251
0.0168	20.6897	1200	6.1418	0.3448	0.3309
0.0168	22.4138	1300	6.2330	0.3417	0.3281
0.0168	24.1379	1400	6.2520	0.3474	0.3344
0.0025	25.8621	1500	6.3587	0.3377	0.3133
0.0025	27.5862	1600	6.3407	0.3408	0.3241
0.0025	29.3103	1700	6.3701	0.3404	0.3215

Framework versions

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