distilbert-base-uncased-finetuned-emotion

This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.3872
• Accuracy: 0.6802
• F1: 0.6793

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	48	1.7238	0.6762	0.6813
0.0305	2.0	96	1.8028	0.6775	0.6755
0.0305	3.0	144	1.9018	0.6689	0.6668
0.0257	4.0	192	1.9426	0.6735	0.6740
0.0257	5.0	240	1.9829	0.6662	0.6670
0.0207	6.0	288	1.9462	0.6722	0.6753
0.0207	7.0	336	1.9573	0.6861	0.6851
0.0185	8.0	384	2.0147	0.6808	0.6820
0.0185	9.0	432	2.0982	0.6669	0.6649
0.0172	10.0	480	2.0431	0.6815	0.6799
0.0172	11.0	528	2.0935	0.6768	0.6751
0.0182	12.0	576	2.0599	0.6868	0.6835
0.0182	13.0	624	2.0953	0.6808	0.6812
0.0148	14.0	672	2.1115	0.6788	0.6790
0.0148	15.0	720	2.1529	0.6735	0.6765
0.0171	16.0	768	2.1873	0.6702	0.6720
0.0171	17.0	816	2.1534	0.6782	0.6793
0.0142	18.0	864	2.1803	0.6782	0.6773
0.0142	19.0	912	2.2252	0.6802	0.6801
0.0168	20.0	960	2.2221	0.6749	0.6764
0.0168	21.0	1008	2.2365	0.6821	0.6817
0.015	22.0	1056	2.2812	0.6742	0.6728
0.015	23.0	1104	2.2447	0.6729	0.6707
0.0145	24.0	1152	2.3272	0.6709	0.6700
0.0145	25.0	1200	2.2630	0.6788	0.6809
0.0151	26.0	1248	2.2751	0.6808	0.6811
0.0151	27.0	1296	2.3018	0.6768	0.6776
0.0144	28.0	1344	2.3544	0.6676	0.6681
0.0144	29.0	1392	2.3109	0.6821	0.6828
0.0126	30.0	1440	2.3234	0.6795	0.6786
0.0126	31.0	1488	2.3294	0.6755	0.6750
0.0142	32.0	1536	2.3183	0.6875	0.6886
0.0142	33.0	1584	2.2949	0.6808	0.6823
0.0131	34.0	1632	2.3451	0.6788	0.6773
0.0131	35.0	1680	2.3160	0.6828	0.6841
0.0143	36.0	1728	2.3251	0.6828	0.6815
0.0143	37.0	1776	2.4003	0.6762	0.6753
0.0116	38.0	1824	2.3675	0.6775	0.6770
0.0116	39.0	1872	2.3700	0.6749	0.6735
0.0126	40.0	1920	2.3700	0.6841	0.6831
0.0126	41.0	1968	2.3818	0.6795	0.6793
0.0115	42.0	2016	2.3518	0.6815	0.6814
0.0115	43.0	2064	2.3829	0.6802	0.6790
0.0135	44.0	2112	2.3638	0.6782	0.6775
0.0135	45.0	2160	2.3568	0.6775	0.6768
0.0146	46.0	2208	2.3633	0.6788	0.6784
0.0118	47.0	2256	2.3725	0.6788	0.6782
0.0118	48.0	2304	2.3875	0.6815	0.6806
0.0116	49.0	2352	2.3862	0.6795	0.6787
0.0116	50.0	2400	2.3872	0.6802	0.6793

Framework versions

• Transformers 4.41.1
• Pytorch 2.1.2
• Datasets 2.19.1
• Tokenizers 0.19.1