CamemBERT
Collection
Based on Metas's RoBERTa model released in 2019, trained on 138GB of French text.
•
3 items
•
Updated
camembert-base-mrpc
This model is a fine-tuned version of camembert-base on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.4286
- Accuracy: 0.8505
- F1: 0.8928
- Combined Score: 0.8716
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
Training results
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu102
- Datasets 2.1.0
- Tokenizers 0.11.6
- Downloads last month
- 5
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Dataset used to train Intel/camembert-base-mrpc
Collection including Intel/camembert-base-mrpc
Evaluation results
- Accuracy on GLUE MRPCself-reported0.850
- F1 on GLUE MRPCself-reported0.893
- Accuracy on gluevalidation set verified0.850
- Precision on gluevalidation set verified0.876
- Recall on gluevalidation set verified0.910
- AUC on gluevalidation set verified0.903
- F1 on gluevalidation set verified0.893
- loss on gluevalidation set verified0.429