• A very simple and unreliable model to score a sentence if it looks classic (Old Arabic) or not.

Usage

import evaluate

# Load the metric
fluency = evaluate.load("Baleegh/Fluency_Score")
fluency.device # 'cuda'

# Some example sentences
texts = ['هلا أعطيتني هذا', 'هل تستطيع أن تعطيني هذا']

# Compute the scores
scores = fluency.compute(texts=texts)['classical_score']
'''array([0.95677805, 0.01825629], dtype=float32)'''

for i in range(len(texts)):
    print(f"{texts[i]} : Fluency {scores[i]:.2f}")

"""
هلا أعطيتني هذا : Fluency 0.96
هل تستطيع أن تعطيني هذا : Fluency 0.02

"""

تنوية

قد لا ينفع هذا المقياس مع نصوصك حتى وإن كانت فصيحة، جربه جيدا.

Downloads last month
74
Safetensors
Model size
109M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.

Model tree for Baleegh/Fluency_Score

Finetuned
(5)
this model

Space using Baleegh/Fluency_Score 1