Update app.py

app.py

@@ -2,22 +2,39 @@ import os
 import yaml
 import gdown 
 import gradio as gr
-from predict import PredictTri
+from predict import PredictTri  # تأكد من وجود ملف predict.py في نفس المجلد
 from huggingface_hub import hf_hub_download
 
+# --- تنزيل ملفات النموذج المطلوبة إذا لم تكن موجودة ---
+
+# تنزيل النموذج الأول من Google Drive
 output_path = "tashkeela-d2.pt"
 gdrive_templ = "https://drive.google.com/file/d/{}/view?usp=sharing"
 if not os.path.exists(output_path):
+    print(f"Downloading model file: {output_path}...")
     model_gdrive_id = "1FGelqImFkESbTyRsx_elkKIOZ9VbhRuo"
     gdown.download(gdrive_templ.format(model_gdrive_id), output=output_path, quiet=False, fuzzy=True)
+    print("Download complete.")
 
+# تنزيل ملف المفردات (vocab) من Google Drive
 output_path = "vocab.vec"
 if not os.path.exists(output_path):
+    print(f"Downloading vocab file: {output_path}...")
     vocab_gdrive_id = "1-0muGvcSYEf8RAVRcwXay4MRex6kmCii"
     gdown.download(gdrive_templ.format(vocab_gdrive_id), output=output_path, quiet=False, fuzzy=True)
+    print("Download complete.")
 
+# تنزيل النموذج الثاني من Hugging Face Hub
 if not os.path.exists("td2/tashkeela-ashaar-td2.pt"):
+    print("Downloading TD2 model from Hugging Face Hub...")
     hf_hub_download(repo_id="munael/Partial-Arabic-Diacritization-TD2", filename="tashkeela-ashaar-td2.pt", local_dir="td2")
+    print("Download complete.")
+
+# --- تحميل الإعدادات وتهيئة المتنبئ ---
+
+# تأكد من وجود ملف config.yaml
+if not os.path.exists("config.yaml"):
+    raise FileNotFoundError("Error: config.yaml file not found. Please create it with the required configurations.")
 
 with open("config.yaml", 'r', encoding="utf-8") as file:
     config = yaml.load(file, Loader=yaml.FullLoader)
@@ -30,6 +47,8 @@ predictor = PredictTri(config)
 current_model_name = "TD2"
 config["model-name"] = current_model_name
 
+# --- تعريف دوال التشكيل (نقاط النهاية للـ API) ---
+
 def diacritze_full(text, model_name):
     global current_model_name, predictor
     if model_name != current_model_name:
@@ -55,106 +74,54 @@ def diacritze_partial(text, mask_mode, threshold, model_name):
     diacritized_lines = predictor.predict_partial(do_partial=do_partial, lines=text.split('\n'))
     return diacritized_lines
 
+# --- بناء واجهة المستخدم باستخدام Gradio ---
+
 with gr.Blocks(theme=gr.themes.Default(text_size="lg")) as demo:
 
     gr.Markdown(
     """
-    <img src='https://huggingface.co/spaces/bkhmsi/Partial-Arabic-Diacritization/resolve/main/PartialDD.png' style='float:right; margin: 0 0 10px 10px; width: 20%'/>
-
-    <h1> Partial Diacritization: A Context-Contrastive Inference Approach </h1>
-    <h2> Authors: Muhammad ElNokrashy, Badr AlKhamissi </h2>
-    <h3> Paper: <a href="https://arxiv.org/abs/2401.08919"> https://arxiv.org/abs/2401.08919 </a> </h3>
-    <details>
-    <summary>Abstract</summary>
-    <p>Diacritization plays a pivotal role in improving readability and disambiguating the meaning of Arabic texts. Efforts have so far focused on marking every eligible character (Full Diacritization). Comparatively overlooked, Partial Diacritzation (PD) is the selection of a subset of characters to be marked to aid comprehension where needed.Research has indicated that excessive diacritic marks can hinder skilled readers---reducing reading speed and accuracy. We conduct a behavioral experiment and show that partially marked text is often easier to read than fully marked text, and sometimes easier than plain text. In this light, we introduce Context-Contrastive Partial Diacritization (CCPD)---a novel approach to PD which integrates seamlessly with existing Arabic diacritization systems. CCPD processes each word twice, once with context and once without, and diacritizes only the characters with disparities between the two inferences. Further, we introduce novel indicators for measuring partial diacritization quality {SR, PDER, HDER, ERE}, essential for establishing this as a machine learning task. Lastly, we introduce TD2, a Transformer-variant of an established model which offers a markedly different performance profile on our proposed indicators compared to all other known systems.</p>
-    </details>
+    <h1>أداة تشكيل النصوص العربية</h1>
+    <p>هذا الخادم يعمل محلياً على جهازك.</p>
     """)
 
-
     model_choice = gr.Dropdown(
         choices=["D2", "TD2"], 
-        label="Diacritization Model", 
+        label="نموذج التشكيل", 
         value=current_model_name
     )
     
-    with gr.Tab(label="Partial Diacritization") as partial_settings:
+    with gr.Tab(label="التشكيل الجزئي") as partial_settings:
         with gr.Row():
-            masking_mode = gr.Radio(choices=["Hard", "Soft"], value="Hard", label="Masking Mode")
-            threshold_slider = gr.Slider(label="Soft Masking Threshold", minimum=0, maximum=1, value=0.1)
+            masking_mode = gr.Radio(choices=["Hard", "Soft"], value="Hard", label="نمط الإخفاء")
+            threshold_slider = gr.Slider(label="عتبة الإخفاء الناعم", minimum=0, maximum=1, value=0.1)
         
         partial_input_txt = gr.Textbox(
-            placeholder="اكتب هنا", 
-            lines=5, 
-            label="Input",
-            type='text',
-            rtl=True,
-            text_align='right',
+            placeholder="اكتب هنا", lines=5, label="النص المدخل", rtl=True
         )
-
         partial_output_txt = gr.Textbox(
-            lines=5, 
-            label="Output",
-            type='text',
-            rtl=True,  
-            text_align='right',
-            show_copy_button=True,
-        )
-
-        partial_btn = gr.Button(value="Shakkel")
-        partial_btn.click(diacritze_partial, inputs=[partial_input_txt, masking_mode, threshold_slider, model_choice], outputs=[partial_output_txt], queue=True)
-
-        gr.Examples(
-            examples=[
-                ["ولو حمل من مجلس الخيار ، ولم يمنع من الكلام", "Hard", 0, "TD2"],
-                ["ولو حمل من مجلس الخيار ، ولم يمنع من الكلام", "Soft", 0.1, "TD2"],
-                ["ولو حمل من مجلس الخيار ، ولم يمنع من الكلام", "Soft", 0.01, "TD2"],
-            ],
-            inputs=[partial_input_txt, masking_mode, threshold_slider, model_choice],
-            outputs=partial_output_txt,
-            fn=diacritze_partial,
-            cache_examples=True,
+            lines=5, label="النص المخرَج", rtl=True, show_copy_button=True
         )
+        partial_btn = gr.Button(value="شَكِّل جزئياً")
+        partial_btn.click(diacritze_partial, inputs=[partial_input_txt, masking_mode, threshold_slider, model_choice], outputs=[partial_output_txt], api_name="diacritze_partial")
 
-    with gr.Tab(label="Full Diacritization"):
-
+    with gr.Tab(label="التشكيل الكامل"):
         full_input_txt = gr.Textbox(
-            placeholder="اكتب هنا", 
-            lines=5, 
-            label="Input",
-            type='text',
-            rtl=True,
-            text_align='right',
+            placeholder="اكتب هنا", lines=5, label="النص المدخل", rtl=True
         )
-
         full_output_txt = gr.Textbox(
-            lines=5, 
-            label="Output",
-            type='text',
-            rtl=True,  
-            text_align='right',
-            show_copy_button=True,
+            lines=5, label="النص المخرَج", rtl=True, show_copy_button=True
         )
+        full_btn = gr.Button(value="شَكِّل بالكامل")
+        full_btn.click(diacritze_full, inputs=[full_input_txt, model_choice], outputs=[full_output_txt], api_name="diacritze_full")
 
-        full_btn = gr.Button(value="Shakkel")
-        full_btn.click(diacritze_full, inputs=[full_input_txt, model_choice], outputs=[full_output_txt], queue=True)
-
-        gr.Examples(
-            examples=[
-                ["ولو حمل من مجلس الخيار ، ولم يمنع من الكلام", "TD2"],
-            ],
-            inputs=[full_input_txt, model_choice],
-            outputs=full_output_txt,
-            fn=diacritze_full,
-            cache_examples=True,
-        )
+# --- تشغيل الخادم (الجزء الذي تم تعديله) ---
 
 if __name__ == "__main__":
+    print("Starting Gradio server on local network...")
+    # هذا السطر يجعل الخادم متاحاً على شبكتك المحلية
+    # استخدم عنوان IP الخاص بجهازك مع المنفذ 7860 للاتصال به من الأجهزة الأخرى
     demo.queue().launch(
-        # share=False,
-        # debug=False,
-        # server_port=7860,
-        # server_name="0.0.0.0",
-        # ssl_verify=False,
-        # ssl_certfile="cert.pem",
-        # ssl_keyfile="key.pem"
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=False
     )