Modelpublic

hw2942/bert-base-chinese-finetuning-wallstreetcn-morning-news-market-overview-open-000001SH-v1

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes8downloads

Model Card

bert-base-chinese-finetuning-wallstreetcn-morning-news-market-overview-open-000001SH-v1

This model is a fine-tuned version of bert-base-chinese on the dataset of Wallstreetcn Morning News Market Overview with overnight index (000001.SH) movement labels. It achieves the following results on the evaluation set:

—Loss: 0.6601119637489319
—Accuracy: 0.7241379310344828

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
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	38	0.6936	0.4483
No log	2.0	76	0.6850	0.5862
No log	3.0	114	0.6977	0.5862
No log	4.0	152	0.6579	0.6207
No log	5.0	190	0.7235	0.4483
No log	6.0	228	0.6601	0.7241
No log	7.0	266	0.6510	0.6897
No log	8.0	304	0.7066	0.7241
No log	9.0	342	0.8716	0.6552
No log	10.0	380	0.8149	0.6207

Framework versions

—Transformers 4.31.0
—Pytorch 2.0.1+cu118
—Datasets 2.14.2
—Tokenizers 0.13.3