finbert-finetuned-FG-SINGLE_SENTENCE-NEWS-WEIGHTED

This model is a fine-tuned version of ProsusAI/finbert on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.2741
• Accuracy: 0.7475
• F1: 0.7253

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: 6e-05
• train_batch_size: 48
• eval_batch_size: 48
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	249	0.9150	0.7346	0.6484
No log	2.0	498	0.8837	0.6210	0.6317
1.033	3.0	747	0.8460	0.6485	0.6666
1.033	4.0	996	1.0089	0.6831	0.6909
0.5642	5.0	1245	1.2507	0.7352	0.7152
0.5642	6.0	1494	1.3241	0.7129	0.7042
0.2078	7.0	1743	1.5163	0.7528	0.7230
0.2078	8.0	1992	1.5818	0.7352	0.7236
0.1108	9.0	2241	1.7930	0.7012	0.7046
0.1108	10.0	2490	1.8262	0.7305	0.7211
0.07	11.0	2739	2.0415	0.7440	0.7192
0.07	12.0	2988	2.1260	0.7563	0.7230
0.0392	13.0	3237	2.1502	0.7528	0.7323
0.0392	14.0	3486	2.2117	0.7516	0.7270
0.0174	15.0	3735	2.2657	0.7405	0.7236
0.0174	16.0	3984	2.2741	0.7475	0.7253

Framework versions

• Transformers 4.16.2
• Pytorch 1.9.1
• Datasets 1.18.4
• Tokenizers 0.11.6