Modelpublic

jayantapaul888/twitter-data-prosusai-finbert-sentiment-finetuned-memes

sourceHugging Faceupdated 4y agoView on Hugging Face

0likes12downloads

Model Card

twitter-data-prosusai-finbert-sentiment-finetuned-memes

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

—Loss: 0.5328
—Accuracy: 0.9274
—Precision: 0.9277
—Recall: 0.9274
—F1: 0.9275

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
—trainbatchsize: 64
—evalbatchsize: 64
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.3284	1.0	1783	0.2770	0.9140	0.9152	0.9140	0.9140
0.2422	2.0	3566	0.2388	0.9282	0.9297	0.9282	0.9283
0.207	3.0	5349	0.2291	0.9325	0.9337	0.9325	0.9326
0.1829	4.0	7132	0.2298	0.9341	0.9347	0.9341	0.9340
0.1665	5.0	8915	0.2493	0.9303	0.9309	0.9303	0.9304
0.142	6.0	10698	0.2749	0.9306	0.9309	0.9306	0.9305
0.1188	7.0	12481	0.2687	0.9279	0.9286	0.9279	0.9280
0.1013	8.0	14264	0.2900	0.9270	0.9273	0.9270	0.9271
0.0849	9.0	16047	0.3247	0.9261	0.9261	0.9261	0.9261
0.0723	10.0	17830	0.3549	0.9259	0.9266	0.9259	0.9261
0.057	11.0	19613	0.3706	0.9283	0.9288	0.9283	0.9284
0.0496	12.0	21396	0.4070	0.9258	0.9266	0.9258	0.9260
0.0423	13.0	23179	0.4361	0.9254	0.9262	0.9254	0.9256
0.0355	14.0	24962	0.4602	0.9297	0.9305	0.9297	0.9298
0.0291	15.0	26745	0.4859	0.9258	0.9259	0.9258	0.9258
0.0248	16.0	28528	0.5024	0.9273	0.9274	0.9273	0.9273
0.0219	17.0	30311	0.5093	0.9263	0.9264	0.9263	0.9263
0.0191	18.0	32094	0.5244	0.9280	0.9288	0.9280	0.9282
0.0171	19.0	33877	0.5290	0.9273	0.9276	0.9273	0.9274
0.0172	20.0	35660	0.5328	0.9274	0.9277	0.9274	0.9275

Framework versions

—Transformers 4.24.0.dev0
—Pytorch 1.11.0+cu102
—Datasets 2.6.1
—Tokenizers 0.13.1