ternary_persian_sentiment_analysis

This model is a fine-tuned version of sentence-transformers/LaBSE on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4914
• Accuracy: 0.8458
• F1 Score: 0.8459

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: 1e-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	F1 Score
0.496	1.0	1394	0.4708	0.8192	0.8186
0.4177	2.0	2788	0.4914	0.8458	0.8459
0.3237	3.0	4182	0.5736	0.8354	0.8356
0.2874	4.0	5576	0.7309	0.8216	0.8217
0.2142	5.0	6970	0.9256	0.8184	0.8187
0.1494	6.0	8364	1.0608	0.8200	0.8197
0.1396	7.0	9758	1.0638	0.8257	0.8256
0.0983	8.0	11152	1.2088	0.8200	0.8200
0.0907	9.0	12546	1.3653	0.8079	0.8083
0.0727	10.0	13940	1.3032	0.8305	0.8307
0.0526	11.0	15334	1.4689	0.8184	0.8184
0.0413	12.0	16728	1.4875	0.8224	0.8227
0.0265	13.0	18122	1.6185	0.8241	0.8242
0.0366	14.0	19516	1.7135	0.8168	0.8168
0.0251	15.0	20910	1.6607	0.8249	0.8251
0.0151	16.0	22304	1.7922	0.8111	0.8111
0.0159	17.0	23698	1.7303	0.8200	0.8201
0.0163	18.0	25092	1.7555	0.8232	0.8232
0.0178	19.0	26486	1.7680	0.8232	0.8232
0.0147	20.0	27880	1.7546	0.8216	0.8217

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0