Erlangshen-Longformer-110M-wallstreetcn-morning-news-market-overview-SSE50-f1-12

This model is a fine-tuned version of IDEA-CCNL/Erlangshen-Longformer-110M on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5817
• F1: 0.6364

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	F1
No log	1.0	34	0.7556	0.4211
No log	2.0	68	0.9109	0.6154
No log	3.0	102	1.1187	0.7333
No log	4.0	136	1.5897	0.6154
No log	5.0	170	1.3606	0.5714
No log	6.0	204	1.5822	0.5455
No log	7.0	238	1.9276	0.3529
No log	8.0	272	1.4752	0.64
No log	9.0	306	1.5805	0.5714
No log	10.0	340	1.5817	0.6364

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1