finance_news_classifier

This model is a fine-tuned version of cardiffnlp/twitter-xlm-roberta-base-sentiment on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1719
• Accuracy: 0.8680

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: 16
• eval_batch_size: 16
• 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
No log	1.0	243	0.4023	0.8412
No log	2.0	486	0.4435	0.8526
0.3668	3.0	729	0.5688	0.8402
0.3668	4.0	972	0.6626	0.8598
0.1479	5.0	1215	0.8238	0.8557
0.1479	6.0	1458	0.9073	0.8536
0.0654	7.0	1701	0.9993	0.8557
0.0654	8.0	1944	1.0495	0.8526
0.0368	9.0	2187	1.1007	0.8392
0.0368	10.0	2430	1.1122	0.8505
0.0212	11.0	2673	1.1024	0.8680
0.0212	12.0	2916	1.0697	0.8670
0.0148	13.0	3159	1.1283	0.8639
0.0148	14.0	3402	1.1176	0.8701
0.008	15.0	3645	1.1625	0.8660
0.008	16.0	3888	1.1794	0.8639
0.0052	17.0	4131	1.1701	0.8629
0.0052	18.0	4374	1.1919	0.8608
0.005	19.0	4617	1.1745	0.8670
0.005	20.0	4860	1.1719	0.8680

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3