aift-model-review-multiple-label-classification

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.0090
• Accuracy Thresh: 0.9179

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Thresh
No log	1.0	129	2.2623	0.3531
No log	2.0	258	1.6019	0.5913
No log	3.0	387	1.3774	0.7378
2.6175	4.0	516	1.3383	0.7875
2.6175	5.0	645	1.2456	0.8006
2.6175	6.0	774	1.3044	0.8679
2.6175	7.0	903	1.4123	0.8746
0.7127	8.0	1032	1.5500	0.8872
0.7127	9.0	1161	1.6639	0.8894
0.7127	10.0	1290	1.8716	0.9024
0.7127	11.0	1419	1.8131	0.8985
0.3804	12.0	1548	2.1177	0.9059
0.3804	13.0	1677	2.1873	0.9105
0.3804	14.0	1806	2.3237	0.9098
0.3804	15.0	1935	2.5947	0.9112
0.2297	16.0	2064	2.5776	0.9116
0.2297	17.0	2193	2.7601	0.9158
0.2297	18.0	2322	2.6187	0.9165
0.2297	19.0	2451	2.9175	0.9165
0.1588	20.0	2580	2.9085	0.9168
0.1588	21.0	2709	2.8516	0.9183
0.1588	22.0	2838	2.8932	0.9179
0.1588	23.0	2967	2.8514	0.9175
0.1154	24.0	3096	3.0075	0.9179
0.1154	25.0	3225	3.0090	0.9179

Framework versions

• Transformers 4.38.0.dev0
• Pytorch 2.2.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2