使用FLASKRESTAPI的机器学习模型

# importing necessary libraries
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn import metrics# importing dataset from sklearn
from sklearn.datasets import load_boston
boston_data &＃61; load_boston()# initializing dataset
data_ &＃61; pd.DataFrame(boston_data.data)### Top five rows of dataset
data_.head()


# Adding features names to the dataframe
data_.columns &＃61; boston_data.feature_names
data_.head()


# Target feature of Boston Housing data
data_[ &＃39;PRICE&＃39; ] &＃61; boston_data.target


# checking null values
data_.isnull().sum()

# checking if values are categorical or not
data_.info()

# creating feature and target variable
X &＃61; data_.drop([ &＃39;PRICE&＃39; ], axis&＃61; 1 )
y &＃61; data_[ &＃39;PRICE&＃39; ]# splitting into training and testing set
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test &＃61; train_test_split(X,y,test_size&＃61; 0.2 , random_state&＃61; 1 )
print( "X training shape : " , X_train.shape )
print( "X test shape : " , X_test.shape )
print( "y training shape :“ , y_train.shape )
print(" y test shape :”, y_test.shape )# creating model
from sklearn.ensemble import RandomForestRegressor
classifier &＃61; RandomForestRegressor()
classifier.fit(X_train, y_train)

# Model evaluation for training data
prediction &＃61; classifier.predict(X_train)
print( "r^2 : " , metrics.r2_score(y_train, prediction))
print( "Mean Absolute Error: " , metrics.mean_absolute_error(y_train, prediction))
print( "Mean Squared Error: " , metrics.mean_squared_error(y_train, prediction))
print( "Root Mean Squared Error : " , np.sqrt(metrics.mean_squared_error(y_train, prediction)))# Model evaluation for testing data
prediction_test &＃61; classifier.predict(X_test)
print( "r^2 : " , metrics.r2_score(y_test, prediction_test))
print( "Mean Absolute Error : " , metrics.mean_absolute_error(y_test, prediction_test))
print( "Mean Squared Error : " , metrics.mean_squared_error(y_test, prediction_test))
print( "Root Mean Absolute Error : " , np.sqrt(metrics.mean_squared_error(y_test, prediction_test)))


# saving the model
import pickle
with open( &＃39;model/model.pkl&＃39; , &＃39;wb&＃39; ) as file:pickle.dump(classifier, file)# saving the columns
model_columns &＃61; list(X.columns)
with open( &＃39;model/model_columns.pkl&＃39; , &＃39;wb&＃39; ) as file:pickle.dump(model_columns, file)

>> pip install Flask

from flask import Flaskapp &＃61; Flask(__name__)&＃64;app.route(&＃39;/&＃39;, methods&＃61;[&＃39;GET&＃39;, &＃39;POST&＃39;])
def main () :return "Boston House Price Prediction”if __name__ &＃61;&＃61; " __main__ ":app.run()

>> export FLASK_APP&＃61;app.py
>> export FLASK_ENV&＃61;development
>> flask run

from flask import render_template, request, jsonify
import flask
import numpy as np
import traceback
import pickle
import pandas as pd# App definition
app &＃61; Flask(__name__,template_folder&＃61; &＃39;templates&＃39; )# importing models
with open( &＃39;webapp/model/model.pkl&＃39; , &＃39;rb&＃39; ) as f:classifier &＃61; pickle.load (f)with open( &＃39;webapp/model/model_columns.pkl&＃39; , &＃39;rb&＃39; ) as f:model_columns &＃61; pickle.load (f)&＃64;app.route(&＃39;/&＃39;)
def welcome () :return "Boston Housing Price Prediction"&＃64;app.route(&＃39;/predict&＃39;, methods&＃61;[&＃39;POST&＃39;,&＃39;GET&＃39;])
def predict () :if flask.request.method &＃61;&＃61; &＃39;GET&＃39; :return "Prediction page"if flask.request.method &＃61;&＃61; &＃39;POST&＃39; :try :json_ &＃61; request.jsonprint(json_)query_ &＃61; pd.get_dummies(pd.DataFrame(json_))query &＃61; query_.reindex(columns &＃61; model_columns, fill_value&＃61; 0 )prediction &＃61; list(classifier.predict(query))return jsonify({"prediction" :str(prediction)})except :return jsonify({"trace" : traceback.format_exc()})if __name__ &＃61;&＃61; "__main__" :app.run()