Set up environment variable

app.py

@@ -1,88 +1,126 @@
-import gradio as gr
-import time
-from datetime import datetime
-import pandas as pd
-from sentence_transformers import SentenceTransformer
-from qdrant_client import QdrantClient
-from qdrant_client.models import Filter, FieldCondition, MatchValue
-
-# Load model
-model = SentenceTransformer('intfloat/multilingual-e5-small')
-
-qdrant_client = QdrantClient(
-    url=url,
-    api_key=api_key,
-)
-
-# Global cache to hold current query/result
-latest_query_result = {"query": "", "result": ""}
-
-# Feedback logger
-def log_feedback(feedback):
-    now = datetime.now().isoformat()
-    log_entry = {
-        "timestamp": now,
-        "query": latest_query_result["query"],
-        "result": latest_query_result["result"],
-        "feedback": feedback
-    }
-    df = pd.DataFrame([log_entry])
-    df.to_csv("feedback_log.csv", mode='a', header=not pd.io.common.file_exists("feedback_log.csv"), index=False)
-    return f"✅ Feedback saved: {feedback}"
-
-# Main search function
-def search_product(query):
-    if (query == ""):
-      return
-    start_search = time.time()
-    start_embed = time.time()
-    query_embed = model.encode("query: " + query)
-    embed_time = time.time() - start_embed
-
-    start_query = time.time()
-    result = qdrant_client.query_points(
-        collection_name="product_E5",
-        query=query_embed.tolist(),
-        with_payload=True,
-        query_filter=Filter(
-            must=[FieldCondition(key="type", match=MatchValue(value="product"))]
-        )
-    ).points
-    query_time = time.time() - start_query
-    search_time = time.time() - start_search
-
-    output = f"🕐 Embedding Time: {embed_time:.4f} sec\n"
-    output += f"🔍 Query Time: {query_time:.4f} sec\n"
-    output += f"⏱ Total Search Time: {search_time:.4f} sec\n\n"
-    output += "📦 ผลลัพธ์:\n"
-
-    result_summary = ""
-    for res in result:
-        line = f"- {res.payload.get('name', '')} (score: {res.score:.4f})"
-        output += line + "\n"
-        result_summary += line + " | "
-
-    latest_query_result["query"] = query
-    latest_query_result["result"] = result_summary.strip()
-
-    return output
-
-# Gradio UI
-with gr.Blocks() as demo:
-    gr.Markdown("## 🔎 Product Semantic Search (E5 + Qdrant)")
-
-    query_input = gr.Textbox(label="ใส่คำค้นหาสินค้า")
-    result_output = gr.Textbox(label="📋 ผลลัพธ์")
-
-    with gr.Row():
-        match_btn = gr.Button("✅ ตรงกับที่ต้องการ")
-        not_match_btn = gr.Button("❌ ไม่ตรง")
-
-    feedback_status = gr.Textbox(label="📬 สถานะ Feedback")
-
-    query_input.submit(search_product, inputs=query_input, outputs=result_output)
-    match_btn.click(lambda: log_feedback("match"), outputs=feedback_status)
-    not_match_btn.click(lambda: log_feedback("not_match"), outputs=feedback_status)
-
-# Launch
+import gradio as gr
+import time
+from datetime import datetime
+import pandas as pd
+from sentence_transformers import SentenceTransformer
+from qdrant_client import QdrantClient
+from qdrant_client.models import Filter, FieldCondition, MatchValue
+
+import os
+from qdrant_client import QdrantClient
+
+qdrant_client = QdrantClient(
+    url=os.environ.get("Qdrant_url"),
+    api_key=os.environ.get("Qdrant_api")
+)
+
+# โมเดลที่โหลดล่วงหน้า
+models = {
+    "E5 (intfloat/multilingual-e5-small)": SentenceTransformer('intfloat/multilingual-e5-small'),
+    "MiniLM (paraphrase-multilingual-MiniLM-L12-v2)": SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2'),
+    "DistilUSE (distiluse-base-multilingual-cased-v1)": SentenceTransformer('sentence-transformers/distiluse-base-multilingual-cased-v1')
+}
+
+# Global memory to hold feedback state
+latest_query_result = {"query": "", "result": "", "model": ""}
+
+
+# 🔍 Search Functions
+def search_with_e5(query):
+    embed = models["E5 (intfloat/multilingual-e5-small)"].encode("query: " + query)
+    return embed
+
+def search_with_minilm(query):
+    embed = models["MiniLM (paraphrase-multilingual-MiniLM-L12-v2)"].encode(query)
+    return embed
+
+def search_with_distiluse(query):
+    embed = models["DistilUSE (distiluse-base-multilingual-cased-v1)"].encode(query)
+    return embed
+
+
+# 🌟 Main search function
+def search_product(query, model_name):
+    start_time = time.time()
+
+    # Choose encoder function
+    if "E5" in model_name:
+        query_embed = search_with_e5(query)
+    elif "MiniLM" in model_name:
+        query_embed = search_with_minilm(query)
+    elif "DistilUSE" in model_name:
+        query_embed = search_with_distiluse(query)
+    else:
+        return "❌ ไม่พบโมเดล"
+
+    # Query Qdrant
+    result = qdrant_client.query_points(
+        collection_name="product_E5",
+        query=query_embed.tolist(),
+        with_payload=True,
+        query_filter=Filter(
+            must=[FieldCondition(key="type", match=MatchValue(value="product"))]
+        )
+    ).points
+
+    elapsed = time.time() - start_time
+
+    # Format result
+    output = f"⏱ Time: {elapsed:.2f}s\n\n📦 ผลลัพธ์:\n"
+    result_summary = ""
+    for res in result:
+        line = f"- {res.payload.get('name', '')} (score: {res.score:.4f})"
+        output += line + "\n"
+        result_summary += line + " | "
+
+    # Save latest query
+    latest_query_result["query"] = query
+    latest_query_result["result"] = result_summary.strip()
+    latest_query_result["model"] = model_name
+
+    return output
+
+
+# 📝 Logging feedback
+def log_feedback(feedback):
+    now = datetime.now().isoformat()
+    log_entry = {
+        "timestamp": now,
+        "model": latest_query_result["model"],
+        "query": latest_query_result["query"],
+        "result": latest_query_result["result"],
+        "feedback": feedback
+    }
+    df = pd.DataFrame([log_entry])
+    df.to_csv("feedback_log.csv", mode='a', header=not pd.io.common.file_exists("feedback_log.csv"), index=False)
+    return f"✅ Feedback saved: {feedback}"
+
+
+# 🎨 Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("## 🔎 Product Semantic Search (Vector Search + Qdrant)")
+
+    with gr.Row():
+        model_selector = gr.Dropdown(
+            choices=list(models.keys()),
+            label="เลือกโมเดล",
+            value="E5 (intfloat/multilingual-e5-small)"
+        )
+        query_input = gr.Textbox(label="พิมพ์คำค้นหา")
+
+    result_output = gr.Textbox(label="📋 ผลลัพธ์")
+
+    with gr.Row():
+        match_btn = gr.Button("✅ ตรง")
+        not_match_btn = gr.Button("❌ ไม่ตรง")
+
+    feedback_status = gr.Textbox(label="📬 สถานะ Feedback")
+
+    # Events
+    submit_fn = lambda q, m: search_product(q, m)
+    query_input.submit(submit_fn, inputs=[query_input, model_selector], outputs=result_output)
+    match_btn.click(lambda: log_feedback("match"), outputs=feedback_status)
+    not_match_btn.click(lambda: log_feedback("not_match"), outputs=feedback_status)
+
+# Run app
 demo.launch(share=True)