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

This model is a fine-tuned version of fnlp/bart-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: 0.632678747177124
• Accuracy: 0.6551724137931034

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	75	0.6801	0.5862
No log	2.0	150	0.6871	0.5862
No log	3.0	225	0.6725	0.5517
No log	4.0	300	0.6327	0.6552
No log	5.0	375	0.7839	0.5862
No log	6.0	450	0.9481	0.5862
0.6041	7.0	525	1.4396	0.5172
0.6041	8.0	600	1.8405	0.6552
0.6041	9.0	675	2.1651	0.5862
0.6041	10.0	750	2.2611	0.5862

Framework versions

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