metadata

license: cc-by-4.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
base_model: l3cube-pune/hing-roberta
model-index:
  - name: hing-roberta-CM-run-4
    results: []

hing-roberta-CM-run-4

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: 2.5827
Accuracy: 0.7525
Precision: 0.6967
Recall: 0.7004
F1: 0.6980

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
0.8734	1.0	497	0.7673	0.7203	0.6617	0.6600	0.6604
0.6245	2.0	994	0.7004	0.7485	0.6951	0.7137	0.7015
0.4329	3.0	1491	1.0469	0.7223	0.6595	0.6640	0.6538
0.2874	4.0	1988	1.3103	0.7586	0.7064	0.7157	0.7104
0.1837	5.0	2485	1.7916	0.7425	0.6846	0.6880	0.6861
0.1121	6.0	2982	2.0721	0.7465	0.7064	0.7041	0.7003
0.0785	7.0	3479	2.3469	0.7425	0.6898	0.6795	0.6807
0.0609	8.0	3976	2.2775	0.7404	0.6819	0.6881	0.6845
0.0817	9.0	4473	2.1992	0.7686	0.7342	0.7147	0.7166
0.042	10.0	4970	2.2359	0.7565	0.7211	0.7141	0.7106
0.0463	11.0	5467	2.2291	0.7646	0.7189	0.7186	0.7177
0.027	12.0	5964	2.3955	0.7525	0.6994	0.7073	0.7028
0.0314	13.0	6461	2.4256	0.7565	0.7033	0.7153	0.7082
0.0251	14.0	6958	2.4578	0.7565	0.7038	0.7025	0.7027
0.0186	15.0	7455	2.5984	0.7565	0.7141	0.6945	0.6954
0.0107	16.0	7952	2.5068	0.7425	0.6859	0.7016	0.6912
0.0134	17.0	8449	2.5876	0.7606	0.7018	0.7041	0.7029
0.0145	18.0	8946	2.6011	0.7626	0.7072	0.7079	0.7073
0.0108	19.0	9443	2.5861	0.7545	0.6973	0.7017	0.6990
0.0076	20.0	9940	2.5827	0.7525	0.6967	0.7004	0.6980

Framework versions

Transformers 4.20.1
Pytorch 1.10.1+cu111
Datasets 2.3.2
Tokenizers 0.12.1