Update app.py

app.py

@@ -1,18 +1,14 @@
 import gradio as gr
 import numpy as np
-import matplotlib
-matplotlib.use('Agg')
-import matplotlib.pyplot as plt
 from sklearn.metrics.pairwise import cosine_similarity
 from sentence_transformers import SentenceTransformer
 from sklearn.decomposition import PCA
 import plotly.graph_objects as go
-from plotly.subplots import make_subplots
 
-# مدل زبانی برای تحلیل معنایی جملات
+# مدل زبانی
 model = SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2')
 
-# داده‌های مربوط به مکاتب فلسفی
+# داده‌ها
 school_data = {
     "Stoicism": [
         "The key to happiness is accepting things we cannot control.",
@@ -36,7 +32,6 @@ school_data = {
     ]
 }
 
-# داده‌های زمانی و پروفایل روانشناختی
 school_profiles = {
     "Stoicism": {"timeline": "3rd century BCE – 3rd century CE", "profile": "Calm, Resilient, Virtue-centered"},
     "Existentialism": {"timeline": "19th – 20th century", "profile": "Individualistic, Authentic, Anxious"},
@@ -44,73 +39,6 @@ school_profiles = {
     "Empiricism": {"timeline": "17th – 18th century", "profile": "Practical, Observational, Experimental"}
 }
 
-# تابع تحلیل و یافتن مکتب فلسفی
-def analyze_text(text):
-    if not text.strip():
-        return "Please enter a philosophical text.", None, None, None, None, None
-    
-    user_vec = model.encode([text])[0]
-
-    best_school = None
-    best_score = -1
-    best_match = ""
-
-    for school, quotes in school_data.items():
-        for quote in quotes:
-            quote_vec = model.encode([quote])[0]
-            score = cosine_similarity([user_vec], [quote_vec])[0][0]
-            if score > best_score:
-                best_score = score
-                best_school = school
-                best_match = quote
-
-    # گراف مفهومی (Conceptual Map)
-    semantic_plot = create_semantic_plot(user_vec, best_match, best_school)
-
-    # بازگشت خروجی‌ها
-    return (
-        best_school,
-        f"{best_score:.2f}",
-        school_profiles[best_school]["profile"],
-        school_profiles[best_school]["timeline"],
-        best_match,
-        semantic_plot
-    )
-
-# تابع رسم گراف مفهومی
-def create_semantic_plot(user_vec, best_quote, best_school):
-    ref_quotes = school_data[best_school]
-    quote_vecs = model.encode(ref_quotes)
-    labels = [f"Ref {i+1}" for i in range(len(ref_quotes))]
-
-    all_vecs = np.vstack([user_vec, quote_vecs])
-    pca = PCA(n_components=2)
-    reduced_vecs = pca.fit_transform(all_vecs)
-
-    fig = go.Figure()
-    fig.add_trace(go.Scatter(
-        x=[reduced_vecs[0][0]],
-        y=[reduced_vecs[0][1]],
-        mode='markers+text',
-        text=["User Input"],
-        name="User Input",
-        textposition="top center",
-        marker=dict(size=14, symbol='circle', color='blue')
-    ))
-
-    fig.add_trace(go.Scatter(
-        x=reduced_vecs[1:, 0],
-        y=reduced_vecs[1:, 1],
-        mode='markers+text',
-        text=labels,
-        name="Reference Quotes",
-        textposition="top center",
-        marker=dict(size=12, symbol='square', color='orange')
-    ))
-
-    fig.update_layout(title="🧭 Conceptual Map", showlegend=False)
-    return fig
-
 def psychological_analysis(text):
     rules = [
         {
@@ -147,7 +75,6 @@ def psychological_analysis(text):
     
     results = []
     lower_text = text.lower()
-    
     for rule in rules:
         if any(keyword in lower_text for keyword in rule["keywords"]):
             results.append({
@@ -155,19 +82,76 @@ def psychological_analysis(text):
                 "followed": rule["followed"],
                 "ignored": rule["ignored"]
             })
-    
     return results
 
-psychological_results = psychological_analysis(user_input)
+def create_semantic_plot(user_vec, best_school):
+    ref_quotes = school_data[best_school]
+    quote_vecs = model.encode(ref_quotes)
+    labels = [f"Ref {i+1}" for i in range(len(ref_quotes))]
 
-psych_output = ""
-for item in psychological_results:
-    psych_output += f"🧠 Psychological Aspect: {item['aspect']}\n"
-    psych_output += f"✅ If Followed: {item['followed']}\n"
-    psych_output += f"❌ If Ignored: {item['ignored']}\n\n"
+    all_vecs = np.vstack([user_vec, quote_vecs])
+    pca = PCA(n_components=2)
+    reduced_vecs = pca.fit_transform(all_vecs)
+
+    fig = go.Figure()
+    fig.add_trace(go.Scatter(
+        x=[reduced_vecs[0][0]],
+        y=[reduced_vecs[0][1]],
+        mode='markers+text',
+        text=["User Input"],
+        name="User Input",
+        textposition="top center",
+        marker=dict(size=14, symbol='circle', color='blue')
+    ))
+
+    fig.add_trace(go.Scatter(
+        x=reduced_vecs[1:, 0],
+        y=reduced_vecs[1:, 1],
+        mode='markers+text',
+        text=labels,
+        name="Reference Quotes",
+        textposition="top center",
+        marker=dict(size=12, symbol='square', color='orange')
+    ))
+
+    fig.update_layout(title="🧭 Conceptual Map", showlegend=False)
+    return fig
+
+def analyze_text(text):
+    if not text.strip():
+        return "Please enter a philosophical text.", "", "", "", "", None, ""
+
+    user_vec = model.encode([text])[0]
+
+    best_school = None
+    best_score = -1
+    best_match = ""
+
+    for school, quotes in school_data.items():
+        for quote in quotes:
+            quote_vec = model.encode([quote])[0]
+            score = cosine_similarity([user_vec], [quote_vec])[0][0]
+            if score > best_score:
+                best_score = score
+                best_school = school
+                best_match = quote
+
+    semantic_plot = create_semantic_plot(user_vec, best_school)
+    profile = school_profiles[best_school]["profile"]
+    timeline = school_profiles[best_school]["timeline"]
+
+    # روانشناسی
+    psych_results = psychological_analysis(text)
+    psych_output = ""
+    if psych_results:
+        for item in psych_results:
+            psych_output += f"🧠 {item['aspect']}\n✅ If Followed: {item['followed']}\n❌ If Ignored: {item['ignored']}\n\n"
+    else:
+        psych_output = "No clear psychological indicators found."
+
+    return best_school, f"{best_score:.2f}", profile, timeline, best_match, semantic_plot, psych_output
 
 
-# رابط کاربری با Gradio
 with gr.Blocks(title="Philosophical Analyzer") as demo:
     gr.Markdown("## 📝 Enter Philosophical Text")
     input_text = gr.Textbox(lines=4, placeholder="Type or paste a philosophical text...")
@@ -176,26 +160,25 @@ with gr.Blocks(title="Philosophical Analyzer") as demo:
 
     with gr.Row():
         with gr.Column():
-            gr.Markdown("### 🕰️ Philosophy Timeline")
-            timeline = gr.Textbox(label="Timeline")
-
             gr.Markdown("### 🧠 Philosophical School Detected:")
-            school = gr.Textbox(label="School")
+            school = gr.Textbox()
 
             gr.Markdown("### 📈 Similarity Score:")
-            score = gr.Textbox(label="Similarity")
+            score = gr.Textbox()
 
             gr.Markdown("### 🧬 Psychological Profile:")
-            profile = gr.Textbox(label="Profile")
+            profile = gr.Textbox()
+
+            gr.Markdown("### 🕰️ Philosophy Timeline")
+            timeline = gr.Textbox()
+
+            with gr.Accordion("🧾 Most Similar Reference Quote", open=False):
+                best_quote = gr.Textbox()
 
         with gr.Column():
             conceptual_map = gr.Plot(label="🧭 Conceptual Map")
 
-    with gr.Accordion("🧾 Most Similar Reference Quote", open=False):
-        best_quote = gr.Textbox(label="Closest Match")
-
     submit_btn.click(fn=analyze_text, inputs=input_text,
-                     outputs=[school, score, profile, timeline, best_quote, conceptual_map])
+                     outputs=[school, score, profile, timeline, best_quote, conceptual_map, profile])
 
-# اجرا
 demo.launch()