Upload app.py with huggingface_hub

app.py

@@ -1,41 +1,19 @@
-import os
-import shutil
-import zipfile
-import pathlib
+import pathlib, shutil, zipfile
 import pandas
 import gradio
 import huggingface_hub
 import autogluon.tabular
 
-# Settings
 MODEL_REPO_ID = "jennifee/classical_automl_model"
 ZIP_FILENAME  = "autogluon_predictor_dir.zip"
 CACHE_DIR = pathlib.Path("hf_assets")
 EXTRACT_DIR = CACHE_DIR / "predictor_native"
 
-# Feature column names and target column names (these should match the model's expected input)
-FEATURE_COLS_MODEL = [
-    "phone_hours",
-    "computer_hours",
-    "device_count",
-    "sleep_quality",
-    "sleep_time",
-    "sleep_hours",
-]
+FEATURE_COLS_MODEL = ['phone_hours', 'computer_hours', 'device_count', 'sleep_quality', 'sleep_time', 'sleep_hours']
 TARGET_COL = "use_before_bed"
+SLEEP_QUALITY_LABELS = ['good', 'medium', 'bad']
+USE_BEFORE_BED_LABELS = {0: 'No', 1: 'Yes'}
 
-# Encoding for categorical features - these mappings are for the Gradio interface,
-# the model expects the original string/numeric values based on the training data.
-SLEEP_QUALITY_LABELS = ["good", "medium", "bad"]
-
-# Encoding for the target outcome (for displaying the prediction and probability)
-USE_BEFORE_BED_LABELS = {
-    0: "No",
-    1: "Yes",
-}
-
-
-# Download & load the native predictor
 def _prepare_predictor_dir() -> str:
     CACHE_DIR.mkdir(parents=True, exist_ok=True)
     local_zip = huggingface_hub.hf_hub_download(
@@ -57,9 +35,7 @@ def _prepare_predictor_dir() -> str:
 PREDICTOR_DIR = _prepare_predictor_dir()
 PREDICTOR = autogluon.tabular.TabularPredictor.load(PREDICTOR_DIR, require_py_version_match=False)
 
-# A mapping utility to make it easier to encode the variables for display
 def _human_label(c):
-    # Updated to map 0/1 to "No"/"Yes" for the target
     try:
         ci = int(c)
         if ci in USE_BEFORE_BED_LABELS:
@@ -68,25 +44,23 @@ def _human_label(c):
         pass
     return str(c)
 
-# This functions takes all of our features, encodes this accordingly, and performs a prediction
 def do_predict(phone_hours, computer_hours, device_count, sleep_quality_label, sleep_time, sleep_hours):
-    # Create the input row using the original column names and appropriate data types
     row = {
         "phone_hours": float(phone_hours),
         "computer_hours": float(computer_hours),
         "device_count": int(device_count),
-        "sleep_quality": sleep_quality_label, # Pass the label directly
+        "sleep_quality": sleep_quality_label,
         "sleep_time": int(sleep_time),
         "sleep_hours": float(sleep_hours),
     }
     X = pandas.DataFrame([row], columns=FEATURE_COLS_MODEL)
 
-
     pred_series = PREDICTOR.predict(X)
     raw_pred = pred_series.iloc[0]
+    pred_label = _human_label(raw_pred)
 
+    proba = None
     try:
-        # Re-added probability prediction
         proba = PREDICTOR.predict_proba(X)
         if isinstance(proba, pandas.Series):
             proba = proba.to_frame().T
@@ -94,48 +68,36 @@ def do_predict(phone_hours, computer_hours, device_count, sleep_quality_label, s
         print(f"Error calculating probabilities: {e}")
         proba = None
 
-    pred_label = _human_label(raw_pred)
-
     proba_dict = None
     if proba is not None:
         row0 = proba.iloc[0]
         tmp = {}
-        # Assuming the probability columns are the class labels (0 and 1)
         for cls in [0, 1]:
-             if cls in row0:
-                 key = _human_label(cls)
-                 tmp[key] = float(row0[cls]) + float(tmp.get(key, 0.0))
-        proba_dict = dict(sorted(tmp.items(), key=lambda kv: kv[1], reverse=True))
-
+            val = None
+            if cls in row0.index:
+                val = row0[cls]
+            elif str(cls) in row0.index:
+                val = row0[str(cls)]
+            if val is not None:
+                key = _human_label(cls)
+                tmp[key] = float(tmp.get(key, 0.0)) + float(val)
+        if tmp:
+            proba_dict = dict(sorted(tmp.items(), key=lambda kv: kv[1], reverse=True))
 
-    # Updated output format to include confidence
     md = f"**Prediction:** {pred_label}"
     if proba_dict:
-        # Corrected the f-string syntax to be on a single line within the multiline string
         md += f"  \n**Confidence:** {round(proba_dict.get(pred_label, 0.0) * 100, 2)}%"
 
-    # Always return two values, even if proba_dict is None
     return proba_dict, md
 
+EXAMPLES = [[3.5, 5.0, 3, 'good', 23, 7.0], [4.2, 6.5, 3, 'medium', 0, 6.5], [5.0, 4.0, 4, 'bad', 1, 6.0], [2.0, 7.5, 3, 'good', 22, 7.5], [3.8, 6.0, 3, 'medium', 0, 6.0], [4.5, 5.5, 3, 'good', 1, 7.0]]
 
-# Representative examples - Updated examples based on the new dataset
-EXAMPLES = [
-    [3.5, 5.0, 3, "good", 23, 7.0],
-    [4.2, 6.5, 3, "medium", 0, 6.5],
-    [5.0, 4.0, 4, "bad", 1, 6.0],
-    [2.0, 7.5, 3, "good", 22, 7.5],
-    [3.8, 6.0, 3, "medium", 0, 6.0],
-    [4.5, 5.5, 3, "good", 1, 7.0],
-]
-
-# Gradio UI
 with gradio.Blocks() as demo:
-    # Provide an introduction
     gradio.Markdown("# Predict Phone Use Before Bed")
-    gradio.Markdown("""
-This app predicts whether a student uses their phone before bed based on their sleeping habits.
-Enter the student's sleeping habits below to get a prediction.
-""")
+    gradio.Markdown(
+        "This app predicts whether a student uses their phone before bed based on their sleeping habits."
+        "\nEnter the student's sleeping habits below to get a prediction."
+    )
 
     with gradio.Row():
         phone_hours = gradio.Number(value=3.5, precision=1, label=FEATURE_COLS_MODEL[0])
@@ -147,14 +109,10 @@ Enter the student's sleeping habits below to get a prediction.
         sleep_time = gradio.Number(value=23, precision=0, label=FEATURE_COLS_MODEL[4])
         sleep_hours = gradio.Number(value=7.0, precision=1, label=FEATURE_COLS_MODEL[5])
 
-
-    # Re-added Label for probability output and kept Markdown for prediction text
     proba_pretty = gradio.Label(num_top_classes=2, label="Probability of Using Phone Before Bed")
-    prediction_output = gradio.Markdown(label="Prediction")
+    prediction_output = gradio.Markdown()
 
-    # Updated inputs list and outputs list to match the do_predict function
     inputs = [phone_hours, computer_hours, device_count, sleep_quality_label, sleep_time, sleep_hours]
-    # The do_predict function now returns either a dict or a string, update outputs accordingly
     outputs = [proba_pretty, prediction_output]
     for comp in inputs:
         comp.change(fn=do_predict, inputs=inputs, outputs=outputs)
@@ -168,4 +126,4 @@ Enter the student's sleeping habits below to get a prediction.
     )
 
 if __name__ == "__main__":
-    demo.launch(debug=False) # Setting debug back to False for Space
+    demo.launch(debug=False)