Update app.py

app.py

@@ -4,7 +4,11 @@ import torch
 import re
 from typing import Dict
 import textstat
+from fastapi import FastAPI
 
+# =======================
+# إعداد النموذج والمكتبات
+# =======================
 MODEL_PATH = "GhadaAlothman/arabert_readability_3class"
 tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
 model = AutoModelForSequenceClassification.from_pretrained(MODEL_PATH)
@@ -15,39 +19,56 @@ AR_LETTERS = r"[\u0600-\u06FF]"
 SENT_SEP = re.compile(r"[\.!\?؟؛…]+")
 WORD_RE = re.compile(fr"{AR_LETTERS}+")
 
-def strip_diacritics(s: str): 
+
+# =======================
+# دوال المعالجة المساعدة
+# =======================
+def strip_diacritics(s: str):
     return DIACRITICS.sub("", s)
 
-def normalize_arabic(s: str): 
-    return re.sub("[\u0622\u0623\u0625]", "ا", strip_diacritics(s)).replace("ى","ي").replace("ة","ه")
+def normalize_arabic(s: str):
+    return re.sub("[\u0622\u0623\u0625]", "ا", strip_diacritics(s)).replace("ى", "ي").replace("ة", "ه")
 
-def split_sentences(text: str): 
+def split_sentences(text: str):
     return [p.strip() for p in SENT_SEP.split(text) if p.strip()]
 
-def tokenize_words(text: str): 
+def tokenize_words(text: str):
     return WORD_RE.findall(text)
 
-def difficult_word(w: str, min_len: int = 6): 
+def difficult_word(w: str, min_len: int = 6):
     return len(w) >= min_len
-
+    
 def compute_metrics(ar_text: str) -> Dict[str, float]:
     text_norm = normalize_arabic(ar_text)
     sents = split_sentences(text_norm)
     words = tokenize_words(text_norm)
-    n_sents, n_words = max(len(sents),1), max(len(words),1)
+    n_sents, n_words = max(len(sents), 1), max(len(words), 1)
     diff_count = sum(1 for w in words if difficult_word(w))
     try:
         osman = float(textstat.osman(ar_text))
     except Exception:
         osman = 0.0
+
+    # 🔹 حساب عدد الأحرف (بدون مسافات)
+    n_chars = sum(len(w) for w in words)
+
+    # 🔹 حساب مؤشر ARI للعربية
+    ari_ar_score = round((4.71 * (n_chars / n_words)) + (0.5 * (n_words / n_sents)) - 21.43, 3)
+
     return {
         "Word count": n_words,
         "Sentence count": n_sents,
-        "OSMAN_Score": round(osman,3),
+        "Character count": n_chars,
+        "OSMAN_Score": round(osman, 3),
+        "ARI_ArScore": ari_ar_score,
         "Difficult_Words_Count": diff_count,
-        "Average_Sentence_Length_in_Words": round(n_words/n_sents,3)
+        "Average_Sentence_Length_in_Words": round(n_words / n_sents, 3)
     }
 
+
+# =======================
+# دالة التنبؤ بالنص
+# =======================
 def analyze_text(text):
     inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=128)
     with torch.no_grad():
@@ -58,7 +79,13 @@ def analyze_text(text):
     label = label_map[label_id]
     stats = compute_metrics(text)
     return {"Predicted_Label": label, **stats}
-    
+
+
+# =======================
+# واجهة Gradio + FastAPI
+# =======================
+app = FastAPI()
+
 demo = gr.Interface(
     fn=analyze_text,
     inputs=gr.Textbox(label="أدخل النص العربي هنا", lines=6),
@@ -67,6 +94,11 @@ demo = gr.Interface(
     description="أداة ذكية لتقييم مقروئية النصوص العربية باستخدام نموذج AraBERT.",
 )
 
-if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860, share=True)
+# واجهة المستخدم في المسار الرئيسي "/"
+app = gr.mount_gradio_app(app, demo, path="/")
 
+# ✅ واجهة REST API بسيطة في /predict
+@app.post("/predict")
+async def predict(request: dict):
+    text = request.get("text", "")
+    return analyze_text(text)