hing-roberta-CM-run-1

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: 2.4241
• Accuracy: 0.7787
• Precision: 0.7367
• Recall: 0.7378
• F1: 0.7357

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
0.8552	1.0	497	0.6797	0.7103	0.6657	0.6872	0.6648
0.5998	2.0	994	0.6946	0.7304	0.6870	0.7108	0.6933
0.4146	3.0	1491	0.9422	0.7465	0.7215	0.6734	0.6887
0.2592	4.0	1988	1.3122	0.7626	0.7240	0.7130	0.7126
0.1644	5.0	2485	1.7526	0.7344	0.6856	0.6901	0.6875
0.1022	6.0	2982	1.9479	0.7746	0.7331	0.7317	0.7316
0.0764	7.0	3479	2.0772	0.7626	0.7190	0.7214	0.7202
0.0468	8.0	3976	2.2799	0.7626	0.7184	0.7044	0.7099
0.0472	9.0	4473	2.2257	0.7586	0.7103	0.7176	0.7135
0.0306	10.0	4970	2.3307	0.7505	0.7068	0.7081	0.7074
0.0351	11.0	5467	2.2555	0.7666	0.7198	0.7254	0.7219
0.0328	12.0	5964	2.4425	0.7626	0.7258	0.7124	0.7179
0.0225	13.0	6461	2.5229	0.7666	0.7237	0.7138	0.7179
0.0232	14.0	6958	2.5717	0.7646	0.7202	0.7115	0.7144
0.0191	15.0	7455	2.4027	0.7606	0.7110	0.7202	0.7152
0.0175	16.0	7952	2.3918	0.7666	0.7216	0.7241	0.7226
0.0087	17.0	8449	2.4176	0.7767	0.7347	0.7365	0.7345
0.0077	18.0	8946	2.4231	0.7686	0.7201	0.7265	0.7230
0.0095	19.0	9443	2.4162	0.7827	0.7392	0.7436	0.7406
0.0063	20.0	9940	2.4241	0.7787	0.7367	0.7378	0.7357

Framework versions

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