Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
nezha-chinese-base-finetuned-product – AI Model by agdsga | AlphaNeural AI
You can deploy this model and start earning money today!
agdsga
/
nezha-chinese-base-finetuned-product
like
0
transformers
pytorch
tensorboard
bert
text-generation
generated_from_trainer
autotrain_compatible
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
nezha-chinese-base-finetuned-product
This model is a fine-tuned version of
peterchou/nezha-chinese-base
on the None dataset. It achieves the following results on the evaluation set:
Loss: 0.0004
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: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3.0
Training results
Training Loss
Epoch
Step
Validation Loss
0.0309
1.0
6473
0.0037
0.0033
2.0
12946
0.0006
0.0017
3.0
19419
0.0004
Framework versions
Transformers 4.17.0
Pytorch 1.6.0
Datasets 2.0.0
Tokenizers 0.11.6
