finbert-finetuned-FG-SINGLE_SENTENCE-NEWS

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

• Loss: 3.2997
• Accuracy: 0.6414
• F1: 0.6295

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: 32
• eval_batch_size: 32
• 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
No log	1.0	321	0.9371	0.5699	0.4333
0.9282	2.0	642	0.9135	0.5930	0.5447
0.9282	3.0	963	0.9900	0.6033	0.5823
0.6743	4.0	1284	1.0802	0.6142	0.6065
0.3134	5.0	1605	1.5156	0.6183	0.5971
0.3134	6.0	1926	1.3695	0.6319	0.6183
0.1709	7.0	2247	1.8746	0.6462	0.6267
0.1112	8.0	2568	2.0880	0.6176	0.6155
0.1112	9.0	2889	2.3953	0.6190	0.6087
0.0811	10.0	3210	2.3792	0.6339	0.6225
0.0608	11.0	3531	2.3783	0.6360	0.6282
0.0608	12.0	3852	2.5982	0.6544	0.6351
0.039	13.0	4173	2.7687	0.6346	0.6305
0.039	14.0	4494	2.8980	0.6414	0.6299
0.0206	15.0	4815	3.0858	0.6319	0.6253
0.0168	16.0	5136	3.2408	0.6244	0.6170
0.0168	17.0	5457	3.1809	0.6435	0.6293
0.0123	18.0	5778	3.2629	0.6449	0.6324
0.0055	19.0	6099	3.2866	0.6449	0.6308
0.0055	20.0	6420	3.2997	0.6414	0.6295

Framework versions

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