Upload 3 files

app.py

@@ -0,0 +1,38 @@
+# app.py
+import gradio as gr
+import pandas as pd
+import joblib
+
+# --- Step 1: Load the pre-calculated Hype Index data ---
+print("Loading TLD Hype Index data...")
+try:
+    tld_hype_index = joblib.load('tld_hype_index.joblib')
+    # Convert the Series to a DataFrame for better display
+    hype_df = tld_hype_index.reset_index()
+    hype_df.columns = ['TLD', 'Hype Index (0-100)']
+    hype_df['Hype Index (0-100)'] = hype_df['Hype Index (0-100)'].round(2)
+    print("✅ Hype Index data loaded and formatted.")
+except FileNotFoundError:
+    print("❌ ERROR: Data file not found. Make sure 'tld_hype_index.joblib' is uploaded to the Space.")
+    hype_df = pd.DataFrame({'Error': ['tld_hype_index.joblib file not found. Please run the training script and upload the file.']})
+
+
+# --- Step 2: Define the function to display the data ---
+# This function doesn't need any inputs; it just returns the DataFrame we loaded.
+def display_hype_index():
+    return hype_df
+
+# --- Step 3: Create and launch the Gradio web interface ---
+iface = gr.Interface(
+    fn=display_hype_index,
+    inputs=None, # No user input needed for this app
+    outputs=gr.DataFrame(headers=['TLD', 'Hype Index (0-100)'], datatype=['str', 'number']),
+    title="Doma TLD Trend & Hype Analyzer (Model 3)",
+    description="This tool analyzes the recent tokenization activity on the Doma Protocol to generate a 'Hype Index' for Top-Level Domains (TLDs). A higher score indicates more recent momentum and popularity.",
+    live=False, # The app doesn't need to run in real-time
+    allow_flagging="never"
+)
+
+# Launch the app
+if __name__ == "__main__":
+    iface.launch()

requirements.txt

@@ -0,0 +1,7 @@
+requests
+pandas
+scikit-learn==1.3.2
+numpy
+joblib
+gradio
+tqdm

tld_hype_index.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58fae73d917078871c904657fb4fbe75ac58a7043ee0e756a0b57323ce1f4c4a
+size 1320