Update app.py

app.py

@@ -4,162 +4,139 @@ import joblib
 import pandas as pd
 import numpy as np
 import folium
+import sklearn.preprocessing as proc
 import json
 import time
 from datetime import timedelta, datetime
 from branca.element import Figure
 
-from functions import decode_features
+#from functions import decode_features, get_weather_data, get_weather_df, get_weather_json_quick
 #import functions
 
 def greet(total_pred_days):
-    str1 = ""
+
     
-#    print("hi")
+#   print("hi")
     project = hopsworks.login()
-    print("connected")
-#    #api = project.get_dataset_api()
-#    
-#
-#    # The latest available data timestamp
-    start_time = 20221201
-#
-#   # end_time = 1670972400000
-#
-#    #start_date = datetime.now() - timedelta(days=1)
-#    #start_time = int(start_date.timestamp()) * 1000
-#
-#    print("Time Stamp Set. ")
-#    
-#    
-#
-#    print("latest_date")
+#   print("connected")
+#   #api = project.get_dataset_api()
 #   
-#    mr=project.get_model_registry()
-#    model = mr.get_model("temp_model", version=2)
-#    model_dir=model.download()
-#    
-#    model1 = mr.get_model("tempmax_model", version=2) 
-#    model_dir1=model1.download()
-#
-#    model2 = mr.get_model("tempmin_model", version=2)
-#    model_dir2=model2.download()
-#    
-#    model = joblib.load(model_dir + "/model_temp.pkl")
-#    model1 = joblib.load(model_dir1 + "/model_tempmax.pkl")
-#    model2 = joblib.load(model_dir2+ "/model_tempmin.pkl")
-#
-#    print("temp_model is now right")
-#         
-    fs = project.get_feature_store()
-    print("get the store")
-    feature_view = fs.get_feature_view(
-        name = 'weather_fv',
-        version = 1
-    )
-    print("get the fv")
-    X = feature_view.get_batch_data(start_time=start_time)
-    print(X)
-#    print("Data batched")
-#    latest_date_unix = str(X.datetime.values[0])[:10]
-#    latest_date = time.ctime(int(latest_date_unix))
-#    X = X.drop(columns=["datetime"]).fillna(0)
-#    preds = model.predict(X)
-#    preds1= model1.predict(X)
-#    preds2= model2.predict(X)
-#    
-#    
-#
-#   # cities = [city_tuple[0] for city_tuple in cities_coords.keys()]
-#
-#    next_day_date = datetime.today() + timedelta(days=1)
-#    next_day = next_day_date.strftime ('%d/%m/%Y')
-#    str1 = ""
-#    
-#    if(total_pred_days == ""):
-#        return "Empty input"
-#    
-#    count = int(total_pred_days)
-#    if count > 20:
-#        str1 += "Warning: 20 days at most. " + '\n'
-#        count = 20
-#    if count <0:
-#        str1 = "Invalid input."
-#        return str1
-#    
-#    for x in range(count):
-#        if (x != 0):
-#            str1 += (datetime.now() + timedelta(days=x)).strftime('%Y-%m-%d') + " predicted temperature:      " +str(int(preds[len(preds) - count + x]))+ " predicted max temperature:      " +str(int(preds1[len(preds1) - count + x]))+ " predicted min temperature:      " +str(int(preds2[len(preds2) - count + x]))+"\n"
-#    
-#    #print(str1)
-    return str1
 
+#   # The latest available data timestamp
+#   start_time = 1649196000000
 
-demo = gr.Interface(fn=greet, inputs = "text", outputs="text")
-
-
-    
-if __name__ == "__main__":
-    demo.launch()
-
-'''
-def greet(total_pred_days):
-
-    project = hopsworks.login()
-    #api = project.get_dataset_api()
-    fs = project.get_feature_store()
-    feature_view = fs.get_feature_view(
-        name = 'weather_fv',
-        version = 1
-    )
-
-    # The latest available data timestamp
-    start_time = 1635112800000
+#  # end_time = 1670972400000
 
     #start_date = datetime.now() - timedelta(days=1)
     #start_time = int(start_date.timestamp()) * 1000
 
+    #print("Time Stamp Set. ")
 
-    X = feature_view.get_batch_data(start_time=start_time)
-    latest_date_unix = str(X.datetime.values[0])[:10]
-    latest_date = time.ctime(int(latest_date_unix))
-    model = get_model(project=project,
-                      model_name="temp_model",
-                      evaluation_metric="f1_score",
-                      sort_metrics_by="max")
-    model1 = get_model1(project=project,
-                      model_name="tempmax_model",
-                      evaluation_metric="f1_score",
-                      sort_metrics_by="max") 
-    model2 = get_model2(project=project,
-                      model_name="tempmin_model",
-                      evaluation_metric="f1_score",
-                      sort_metrics_by="max")
-    X = X.drop(columns=["datetime"]).fillna(0)
-
-    preds = model.predict(X)
+    #print("latest_date")
     
-    preds1= model1.predict(X)
+    mr=project.get_model_registry()
+    model = mr.get_model("temp_model_new", version=1)
+    model_dir=model.download()
     
-    preds2= model2.predict(X)
+    model1 = mr.get_model("tempmax_model_new", version=1) 
+    model_dir1=model1.download()
+
+    model2 = mr.get_model("tempmin_model_new", version=1)
+    model_dir2=model2.download()
     
+    model = joblib.load(model_dir + "/model_temp_new.pkl")
+    model1 = joblib.load(model_dir1 + "/model_tempmax_new.pkl")
+    model2 = joblib.load(model_dir2+ "/model_tempmin_new.pkl")
 
-   # cities = [city_tuple[0] for city_tuple in cities_coords.keys()]
+    print("temp_model is now right")
+         
+    
+    #X = feature_view.get_batch_data(start_time=start_time)
+    
+    #latest_date_unix = str(X.datetime.values[0])[:10]
+    #latest_date = time.ctime(int(latest_date_unix))
+    # cities = [city_tuple[0] for city_tuple in cities_coords.keys()]
 
-    next_day_date = datetime.today() + timedelta(days=1)
-    next_day = next_day_date.strftime ('%d/%m/%Y')
+    
     str1 = ""
     
     if(total_pred_days == ""):
         return "Empty input"
     
     count = int(total_pred_days)
-    if count > 20:
-        str1 += "Warning: 20 days at most. " + '\n'
-        count = 20
+    if count > 14:
+        str1 += "Warning: 14 days at most. " + '\n'
+        count = 14
     if count <0:
         str1 = "Invalid input."
         return str1
+
+    # Get weather data
+    fs = project.get_feature_store()
+    print("get the store")
+    feature_view = fs.get_feature_view(
+        name = 'weathernew_fv',
+        version = 1
+    )
+    print("get the fv")
+    
+    
+    global X
+    X = pd.DataFrame()
+    
+    for i in range(count+1):
+        # Get, rename column and rescale
+        next_day_date = datetime.today() + timedelta(days=i)
+        next_day = next_day_date.strftime ('%Y-%m-%d')
+        print(next_day)
+        json = get_weather_json_quick(next_day)
+        temp = get_weather_data(json)
+        print("Raw data")
+        print(temp)
+        X = X.append(temp, ignore_index=True)
+    
+    
+    # X reshape
+    
+    X.drop('preciptype', inplace = True, axis = 1)
+    X.drop('severerisk', inplace = True, axis = 1)
+    X.drop('stations', inplace = True, axis = 1)
+    X.drop('sunrise', inplace = True, axis = 1)
+    X.drop('sunset', inplace = True, axis = 1)
+    X.drop('moonphase', inplace = True, axis = 1)
+    X.drop('description', inplace = True, axis = 1)
+    X.drop('icon', inplace = True, axis = 1)
+    
+    X = X.drop(columns=["datetime", "temp", "tempmax", "tempmin", "sunriseEpoch", "sunsetEpoch", "source", "datetimeEpoch", ]).fillna(0) 
+    X = X.rename(columns={'pressure':'sealevelpressure'})
+    X = X.drop(columns = ['conditions'])
+    print("Check dataframe")
+    print(X)
+    print("Data batched.")
+    # Rescale
+    #X = decode_features(X, feature_view=feature_view)
+    # Data scaling
+    #category_cols = ['name','datetime','conditions', 'tempmin', 'tempmax', 'temp']
+
+    #mapping_transformers = {col_name:fs.get_transformation_function(name='standard_scaler') for col_name in col_names if col_name not in category_cols}
+    #category_cols = {col_name:fs.get_transformation_function(name='label_encoder') for col_name in category_cols if col_name not in ['datetime', 'tempmin', 'tempmax', 'temp']}
+
+    #mapping_transformers.update(category_cols)
+    
+    
+    # Data scaling
+    #category_cols = ['conditions']
+    cat_std_cols = ['feelslikemax','feelslikemin','feelslike','dew','humidity','precip','precipprob','precipcover','snow','snowdepth','windgust','windspeed','winddir','sealevelpressure','cloudcover','visibility','solarradiation','solarenergy','uvindex']
+    scaler_std = proc.StandardScaler()
+    #scaler_lb = proc.LabelEncoder()
+    X.insert(19,"conditions",0)
+    X[cat_std_cols] = scaler_std.fit_transform(X[cat_std_cols])
+    #X[category_cols] = scaler_std.transform(X[category_cols])
+    X.insert(0,"name",0)
+    # Predict
+    preds = model.predict(X)
+    preds1= model1.predict(X)
+    preds2= model2.predict(X)
     
     for x in range(count):
         if (x != 0):
@@ -168,16 +145,9 @@ def greet(total_pred_days):
     #print(str1)
     return str1
 
-
-demo = gr.Interface(
-    fn=greet, 
-    inputs=gr.Slider(label="Days of prediction (start from tomorrow)", value=1, minimum=1, maximum=20, step=1),
-    outputs=gr.outputs.Textbox(label="Prediction results"),
-)
+demo = gr.Interface(fn=greet, inputs = "text", outputs="text")
 
 
     
 if __name__ == "__main__":
     demo.launch()
-
-'''