Update app.py

app.py

@@ -5,6 +5,9 @@ import os
 import requests
 from functools import lru_cache
 
+# ----------------------------
+# CONFIG
+# ----------------------------
 JSON_FILE = "form-submissions-1.json"
 MODEL_ID = "HuggingFaceH4/zephyr-7b-beta"
 HF_API_TOKEN = os.environ.get("HF_API_TOKEN")
@@ -31,9 +34,6 @@ CATEGORIES = {
 # ----------------------------
 @lru_cache(maxsize=512)
 def call_zephyr_cached(candidate_str, category_name, job_titles_tuple):
-    """
-    Cached Zephyr LLM call.
-    """
     try:
         prompt = f"""
 You are an HR assistant. Review this candidate and determine if they are suitable for the category '{category_name}'.
@@ -43,10 +43,7 @@ Candidate JSON: {candidate_str}
 
 Respond only 'Yes' if suitable, otherwise 'No'.
 """
-        headers = {
-            "Authorization": f"Bearer {HF_API_TOKEN}",
-            "Content-Type": "application/json"
-        }
+        headers = {"Authorization": f"Bearer {HF_API_TOKEN}", "Content-Type": "application/json"}
         payload = {"inputs": prompt}
         response = requests.post(
             f"https://api-inference.huggingface.co/models/{MODEL_ID}",
@@ -110,31 +107,43 @@ def get_top_candidates(category_name, job_titles, top_n=5):
     df = df.sort_values("Salary_sort").drop(columns=["Salary_sort"])
     return df.head(top_n)
 
+# ----------------------------
+# Show first 5 candidates from raw JSON
+# ----------------------------
+def show_first_candidates():
+    data = json.load(open(JSON_FILE, encoding="utf-8"))
+    first_5 = data[:5]
+    df = pd.DataFrame(first_5)
+    return df
+
 # ----------------------------
 # Gradio interface
 # ----------------------------
 def run_dashboard(category):
     if category not in CATEGORIES:
-        return pd.DataFrame()
-    df = get_top_candidates(category, CATEGORIES[category], top_n=5)
-    return df
-
-def download_csv(category):
+        return pd.DataFrame(), None
     df = get_top_candidates(category, CATEGORIES[category], top_n=5)
     if df.empty:
-        return None
+        return df, None
     file_path = "/tmp/outputs.csv"
     df.to_csv(file_path, index=False)
-    return file_path
+    return df, file_path
 
 demo = gr.Interface(
     fn=run_dashboard,
     inputs=gr.Dropdown(list(CATEGORIES.keys()), label="Select Category"),
     outputs=[gr.Dataframe(label="Top 5 Recommended Candidates"),
-             gr.File(label="Download CSV", file_types=[".csv"], file_path_func=download_csv)],
+             gr.File(label="Download CSV")],
     title="Startup Candidate Dashboard - Zephyr-7B-Beta",
-    description="Top 5 candidates per category using Zephyr LLM with caching. You can download the CSV."
+    description="Top 5 candidates per category using Zephyr LLM. Download CSV available."
 )
 
+# Add separate interface to show first 5 raw candidates
+with gr.Blocks() as app:
+    gr.Markdown("### Raw JSON Preview: First 5 Candidates")
+    gr.Dataframe(show_first_candidates(), label="First 5 Candidates from JSON")
+    gr.Markdown("---")
+    demo.render()
+
 if __name__ == "__main__":
-    demo.launch()
+    app.launch()