Edit model card

darija_test

This model is a fine-tuned version of SI2M-Lab/DarijaBERT on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6520
  • Accuracy: 0.7041

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: 0.0003
  • 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
  • lr_scheduler_warmup_ratio: 0.05
  • num_epochs: 10
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.813 1.0 169 0.7390 0.2959
0.9437 2.0 338 0.6520 0.7041
0.759 3.0 507 0.6080 0.7041
0.7936 4.0 676 0.6582 0.7041
0.8051 5.0 845 0.6452 0.7041
0.7397 6.0 1014 0.6422 0.7041
0.5792 7.0 1183 0.6132 0.7041
0.5339 8.0 1352 0.6095 0.7041
0.5152 9.0 1521 0.6136 0.7041
0.4909 10.0 1690 0.6181 0.7041

Framework versions

  • Transformers 4.42.4
  • Pytorch 2.3.1+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
4
Safetensors
Model size
147M params
Tensor type
F32
·
Inference Examples
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.

Model tree for smerchi/darija_test

Finetuned
(6)
this model