distilbert-base-uncased-finetuned-emotion_new

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: 0.8847
• Accuracy: 0.8
• F1: 0.7333

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: 6
• eval_batch_size: 6
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.5408	1.0	4	0.7674	0.8	0.7333
0.4368	2.0	8	0.7471	0.8	0.7333
0.3222	3.0	12	0.7318	0.8	0.7333
0.4061	4.0	16	0.7289	0.8	0.7333
0.3774	5.0	20	0.7732	0.8	0.7333
0.3304	6.0	24	0.7874	0.8	0.7333
0.3042	7.0	28	0.8036	0.8	0.7333
0.4571	8.0	32	0.8038	0.8	0.7333
0.1992	9.0	36	0.8271	0.8	0.7333
0.2661	10.0	40	0.8498	0.8	0.7333
0.2361	11.0	44	0.8582	0.8	0.7333
0.2292	12.0	48	0.8620	0.8	0.7333
0.2363	13.0	52	0.8678	0.8	0.7333
0.2574	14.0	56	0.8672	0.8	0.7333
0.5177	15.0	60	0.8668	0.8	0.7333
0.226	16.0	64	0.8726	0.8	0.7333
0.1726	17.0	68	0.8788	0.8	0.7333
0.2439	18.0	72	0.8823	0.8	0.7333
0.2005	19.0	76	0.8842	0.8	0.7333
0.2541	20.0	80	0.8847	0.8	0.7333

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu118
• Datasets 2.17.0
• Tokenizers 0.15.2