Fill-Mask
Transformers
Safetensors
PyTorch
Arabic
bert
arabic
maghrebi-arabic
darija
arabizi
code-switching
algeria
morocco
tunisia
libya
masked-language-modeling
Instructions to use NumidiaNLP/MagharibiBERT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NumidiaNLP/MagharibiBERT with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="NumidiaNLP/MagharibiBERT")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("NumidiaNLP/MagharibiBERT") model = AutoModelForMaskedLM.from_pretrained("NumidiaNLP/MagharibiBERT", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download validation_metrics.json from NumidiaNLP/MagharibiBERT: direct link, hf CLI and curl.
- Browser
- Download file 289 Bytes
-
https://huggingface.co/NumidiaNLP/MagharibiBERT/resolve/main/validation_metrics.json
- Command line
-
hf download hf://NumidiaNLP/MagharibiBERT/validation_metrics.json
-
curl -L -o validation_metrics.json https://huggingface.co/NumidiaNLP/MagharibiBERT/resolve/main/validation_metrics.json
289 Bytes
| { | |
| "step": 120000, | |
| "mlm_loss": 2.8314573731724013, | |
| "masked_token_accuracy": 0.519152562066695, | |
| "masked_token_perplexity": 16.970174688827857, | |
| "masked_targets": 129095.0, | |
| "v1_baseline_loss": 3.8699108042074832, | |
| "v1_baseline_accuracy": 0.4071342809558852 | |
| } |