Gherbal
FastText-based multilingual language identification with HuggingFace Hub integration.
Supports 200+ languages including fine-grained Arabic dialect detection.
Installation
pip install gherbal
Quick Start
from gherbal import Gherbal
# Load from HuggingFace Hub
model = Gherbal.from_pretrained("omarkamali/gherbal")
# Predict language
model.predict("Hello, how are you?")
# => [('eng_Latn', 0.99)]
model.predict("مرحبا كيف حالك")
# => [('arb_Arab', 0.95)]
Loading a Local Model
model = Gherbal.from_pretrained("./path/to/model")
Training
import pandas as pd
from gherbal import Gherbal
df = pd.DataFrame({"text": [...], "label": [...]})
model = Gherbal.train(df, save_path="./my_model")
Preprocessing
from gherbal import preprocess_text, create_clean_script_function
text = preprocess_text("Hello @user https://example.com 🎉")
# => "hello"
# Script-aware cleaning
clean = create_clean_script_function()
clean("Latn", "Hello World 123")
# => "Hello World"
Pushing to HuggingFace Hub
model.push_to_hub("username/my-gherbal-model")
License
MIT
Metadata
Release files for gherbal 1.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| gherbal-1.0.1.tar.gz | 21.7 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| gherbal-1.0.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 35.1 kB
Release files / gherbal-1.0.1.tar.gz
| Download URL | gherbal-1.0.1.tar.gz |
|---|---|
| Size | 21.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0cff6694f84b11315ead028e965f1e46cf59fb9ea64bafed5788e07f15b84b2c
|
|
BLAKE2b-256 checksum How to use checksums |
45e1c8f3b2d1fa8c0a6c063259a22df9c4faf22b1e84d3b1912d11a005521d28
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Mar 14, 2026.
Transparency logRelease files / gherbal-1.0.1-py3-none-any.whl
| Download URL | gherbal-1.0.1-py3-none-any.whl |
|---|---|
| Size | 13.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
c8e1b47fc3a697d9bc54dcaa07fab18101595a14fe1f55ebfac0c62953587cb9
|
|
BLAKE2b-256 checksum How to use checksums |
15162ef3e70254d50ba63aebcc4d096b0723f0e84261560e7dd1875c6bea90bb
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Mar 14, 2026.
Transparency log