GISchat-weibo-100k-fine-tuned-bert – AI Model by wsqstar | AlphaNeural AI

You can deploy this model and start earning money today!

wsqstar/
GISchat-weibo-100k-fine-tuned-bert

like

0

text-classification

generated_from_trainer

dirtycomputer/weibo_senti_100k

google-bert/bert-base-chinese

10.57967/hf/2866

autotrain_compatible

endpoints_compatible

Views

No views yet

GISchat-weibo-100k-fine-tuned-bert

This model is a fine-tuned version of bert-base-chinese on weibo-100k dataset.

Github repo: https://github.com/GISChat/Fine-tune-bert

It achieves the following results on the evaluation set:

Loss: 0.0458
Accuracy: 0.9867

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
train_batch_size: 8
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.08	100	0.6573	0.606
0.647	0.16	200	0.2447	0.9507
0.647	0.24	300	0.0914	0.9807
0.1276	0.32	400	0.0609	0.9843
0.1276	0.4	500	0.0607	0.9843
0.0921	0.48	600	0.1053	0.98
0.0921	0.56	700	0.0487	0.9853
0.0885	0.64	800	0.0523	0.9853
0.0885	0.72	900	0.0484	0.986
0.0579	0.8	1000	0.0549	0.985
0.0579	0.88	1100	0.0495	0.9867
0.0507	0.96	1200	0.0458	0.9867

Framework versions

Transformers 4.42.4
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1