hing-roberta-CM-run-5

This model is a fine-tuned version of l3cube-pune/hing-roberta on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.6447
• Accuracy: 0.7525
• Precision: 0.7030
• Recall: 0.7120
• F1: 0.7064

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: 3e-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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.9492	1.0	497	0.7476	0.6157	0.6060	0.6070	0.5171
0.7013	2.0	994	0.7093	0.6982	0.6716	0.6864	0.6663
0.4871	3.0	1491	0.8294	0.7284	0.6714	0.6867	0.6723
0.3838	4.0	1988	1.1275	0.7505	0.6969	0.7025	0.6994
0.254	5.0	2485	1.3831	0.7264	0.6781	0.6975	0.6850
0.1765	6.0	2982	2.0625	0.7384	0.7068	0.6948	0.6896
0.1127	7.0	3479	1.9691	0.7425	0.6925	0.7065	0.6982
0.0757	8.0	3976	2.3871	0.7425	0.7183	0.6926	0.6924
0.0572	9.0	4473	2.4037	0.7344	0.6916	0.6929	0.6882
0.0458	10.0	4970	2.3062	0.7586	0.7174	0.7219	0.7164
0.0405	11.0	5467	2.5591	0.7445	0.6925	0.6964	0.6942
0.0292	12.0	5964	2.5215	0.7384	0.6875	0.6998	0.6917
0.0264	13.0	6461	2.7551	0.7586	0.7122	0.7035	0.7037
0.0299	14.0	6958	2.6536	0.7465	0.7114	0.7088	0.7035
0.0208	15.0	7455	2.5190	0.7505	0.6989	0.7083	0.7030
0.0263	16.0	7952	2.7092	0.7485	0.7076	0.6998	0.6962
0.0077	17.0	8449	2.5933	0.7525	0.7042	0.7143	0.7081
0.009	18.0	8946	2.5831	0.7485	0.6991	0.7152	0.7050
0.0108	19.0	9443	2.6360	0.7545	0.7050	0.7167	0.7098
0.0077	20.0	9940	2.6447	0.7525	0.7030	0.7120	0.7064

Framework versions

• Transformers 4.20.1
• Pytorch 1.10.1+cu111
• Datasets 2.3.2
• Tokenizers 0.12.1