Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
IceBERT-finetuned-grouped – AI Model by ueb1 | AlphaNeural AI
You can deploy this model and start earning money today!
ueb1
/
IceBERT-finetuned-grouped
like
0
transformers
pytorch
tensorboard
roberta
text-classification
generated_from_trainer
gpl-3.0
autotrain_compatible
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
IceBERT-finetuned-grouped
This model is a fine-tuned version of
vesteinn/IceBERT
on an unknown dataset. It achieves the following results on the evaluation set:
Loss: 3.5660
Accuracy: 0.2259
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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5
Training results
Training Loss
Epoch
Step
Validation Loss
Accuracy
No log
1.0
269
4.1727
0.1172
4.3535
2.0
538
3.8406
0.1632
4.3535
3.0
807
3.6718
0.2113
3.6711
4.0
1076
3.5660
0.2259
3.6711
5.0
1345
3.5332
0.2176
Framework versions
Transformers 4.12.5
Pytorch 1.10.0+cu111
Datasets 1.15.1
Tokenizers 0.10.3
