# Step 6.1: Define different input components
import gradio as gr

# a. define text data type
input_module1 = gr.inputs.Slider(1, 100, step=5, label = "Longitude")

# b. define image data type
input_module2 = gr.inputs.Slider(1, 100, step=5, label = "Latitude")

# c. define Number data type
input_module3 = gr.inputs.Slider(1, 100, step=5, label = "Feature3")

# d. define Slider data type
input_module4 = gr.inputs.Slider(1, 100, step=5, label = "Feature4")

# e. define Checkbox data type
input_module5 = gr.inputs.Slider(1, 100, step=5, label = "Feature5")

# f. define Radio data type
input_module6 = gr.inputs.Slider(1, 100, step=5, label = "Feature6")

# g. define Dropdown data type
input_module7 = gr.inputs.Slider(1, 100, step=5, label = "Feature7")

# g. define Dropdown data type
input_module8 = gr.inputs.Slider(1, 100, step=5, label = "Feature8")


# Step 6.2: Define different output components
# a. define text data type
output_module1 = gr.outputs.Textbox(label = "Output Text")

# b. define image data type
output_module2 = gr.outputs.Image(label = "Output Image")

# you can define more output components

def machine_learning_pipeline(input1, input2, input3, input4, input5, input6, input7, input8):
  print("Start ml processing")

  import numpy as np
  import pandas as pd

  print(input1, input2, input3, input4, input5, input6, input7, input8)

  ### 1. process the user submission, collect the features and save them into one numpy array
  new_feature = np.array([[input1, input2, input3, input4, input5, input6, input7, input8]])
  print(new_feature)

  ### 2. follow the data preprocessing steps as we have done in trainig data 
  ### 2.1 check missing values in total_bedroom 

  ### 2.2 feature normalization
  test_set = pd.DataFrame(new_feature, columns = ['longitude', 'latitude', 'housing_median_age', 'total_rooms',
       'total_bedrooms', 'population', 'households', 'median_income'])

  ## 1. clean the missing values in test set
  test_set_clean = test_set.dropna(subset=["total_bedrooms"])
  test_set_clean
  
  ### reload the scaler from the local file
  
  import pickle
  
  with open('minmax_scaler.pkl', 'rb') as f:
      scaler = pickle.load(f)


  ## 4. scale the numeric features in test set. 
  ## important note: do not apply fit function on the test set, using same scalar from training set
  test_features_normalized = scaler.transform(test_set_clean)
  print("test_features_normalized: ",test_features_normalized)

  ### 3. load the pre-train machine learning 


  # load
  with open('tree_reg.pkl', 'rb') as f:
      tree_reg = pickle.load(f)


  ### 4. apply the loaded model on the features to make a prediction 
  ### Step 5: make a prediction using tree model
  test_predictions_trees = tree_reg.predict(test_features_normalized)
  print("test_predictions_trees: ",test_predictions_trees)

  ### 5. send back the prediction 

  print("Start processing")
  import numpy as np
  #output1 = 'This is the output'
  output2 = np.random.rand(28,28)

  import matplotlib.pyplot as plt  
  plt.plot([input1], [input2])
  plt.savefig('test.png')

  return test_predictions_trees, 'test.png'
  
  

gr.Interface(fn=machine_learning_pipeline, 
             inputs = [input_module1, input_module2, input_module3, 
                                          input_module4, input_module5, input_module6,
                                          input_module7, input_module8], 
             outputs = [output_module1, output_module2]).launch(debug=True)