distilbert-base-uncased-finetuned-items-multi-label-21122023

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: 0.2426
• F1 Micro: 0.7434
• F1 Macro: 0.7586
• Hamming Loss: 0.1041
• Jaccard Score: 0.6168

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: 30
• eval_batch_size: 30
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Micro	F1 Macro	Hamming Loss	Jaccard Score
0.2323	1.0	32	0.2750	0.6920	0.6304	0.1188	0.4908
0.1921	2.0	64	0.2568	0.7440	0.7413	0.1001	0.6021
0.1624	3.0	96	0.2522	0.7181	0.7230	0.1121	0.5779
0.1392	4.0	128	0.2443	0.7517	0.7693	0.0988	0.6311
0.1235	5.0	160	0.2416	0.7687	0.7817	0.0948	0.6471
0.1092	6.0	192	0.2374	0.7393	0.7520	0.1055	0.6088
0.0998	7.0	224	0.2386	0.7451	0.7615	0.1041	0.6205
0.0931	8.0	256	0.2424	0.7368	0.7524	0.1068	0.6091
0.0878	9.0	288	0.2435	0.7417	0.7570	0.1041	0.6150
0.087	10.0	320	0.2426	0.7434	0.7586	0.1041	0.6168

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0