upload model

app.py

@@ -78,29 +78,32 @@ def make_clf_t_plot():
     return fig
 
 def download_clf():
-    joblib.dump(clf,"dtc_model.m")
+    # joblib.dump(clf,"dtc_model.m")
     return "./dtc_model.m"
 
 def download_rfc():
-    joblib.dump(rfc,"rfc_model.m")
+    # joblib.dump(rfc,"rfc_model.m")
     return "./rfc_model.m"
 
-table_data, clean_data = make_ra_table(data)
-
-morning_features = ['air_pressure', 'aver_tem', 'humidity',
-       'sunlight_time', 'wind_direction', 'wind_speed']
-feature=clean_data[morning_features].copy()
-label = clean_data['rain_accum'].copy()
-X_train,X_test,y_train,y_test = train_test_split(feature,label,test_size=0.33,random_state=324)
-clf = DecisionTreeClassifier(random_state=25)
-rfc = RandomForestClassifier(random_state=25, n_estimators=11)
-clf.fit(X_train,y_train)
-rfc.fit(X_train,y_train)
-clf_score = clf.score(X_test, y_test)
-rfc_score = rfc.score(X_test, y_test)
-score = pd.DataFrame([[clf_score,rfc_score],['DecisioTree Score','RandomForest Score']],columns=['DecisioTree Score','RandomForest Score'])
+
 
 if __name__ == '__main__':
+
+    table_data, clean_data = make_ra_table(data)
+
+    morning_features = ['air_pressure', 'aver_tem', 'humidity',
+        'sunlight_time', 'wind_direction', 'wind_speed']
+    feature=clean_data[morning_features].copy()
+    label = clean_data['rain_accum'].copy()
+    X_train,X_test,y_train,y_test = train_test_split(feature,label,test_size=0.1,random_state=324)
+    clf = joblib.load("dtc_model.m") # DecisionTreeClassifier(random_state=25)
+    rfc = joblib.load("rfc_model.m") # RandomForestClassifier(random_state=25, n_estimators=11)
+    # clf.fit(X_train,y_train)
+    # rfc.fit(X_train,y_train)
+    clf_score = clf.score(X_test, y_test)
+    rfc_score = rfc.score(X_test, y_test)
+    score = pd.DataFrame([[clf_score,rfc_score],['DecisioTree Score','RandomForest Score']],columns=['DecisioTree Score','RandomForest Score'])
+
     with gr.Blocks() as demo:
         gr.Markdown("""
         ## Data Collection