Skip to main content

deploying Machine learning Model : pkl, Flask,postman

1)Create model and train 

        #  importing Librarys

        import pandas as pd
        import numpy as np
        import matplotlib.pyplot as plt
        import seaborn as sns
        import requests
        from pickle import dump,load

        # Load Dataset

        url="http://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data"

        names = ["sepal_length", "sepal_width", "petal_length", "petal_width", "species"]

        # Loading Dataset
        df=pd.read_csv(url,names=names)

        df.tail(11)

        df.columns
        test=[
        {'sepal_length':5.1, 'sepal_width':3.5, 'petal_length':1.4, 'petal_width':0.2},
            {'sepal_length':6.9, 'sepal_width':3.1, 'petal_length':5.4, 'petal_width':2.1}
        ]

        df.info()

        # Checking null values
        df.isna().sum()

        df.describe()

        # visulization

        plt.figure(figsize=(10,8))
        sns.scatterplot(x='sepal_length',y='sepal_width',data=df,hue='species')

        plt.hist(df['species'])

        # Label Encoding



        # Selecting dependent and indepedent variable
        x=df.drop('species',axis=1)
        y=df['species']

        # split data

        from sklearn.model_selection import train_test_split
        xtrain,xtest,ytrain,ytest=train_test_split(x,y,test_size=0.2,random_state=0)

        # fit and train model

        from sklearn.ensemble import RandomForestClassifier
        model=RandomForestClassifier()
        model.fit(xtrain,ytrain)

        ypred=model.predict(xtest)

        from sklearn.metrics import accuracy_score,classification_report,confusion_matrix
        print("accuray is: ",accuracy_score(ytest,ypred))
        cm=confusion_matrix(ytest,ypred)
        sns.heatmap(cm,annot=True)
        print(classification_report(ytest,ypred))



        # conclusion:
        #In this task we have we explored the iris dataset and predicted the classification using RandomForestClassifier algorithm.

        dump(model,open('model.pkl','wb'))

2) Create Flask app (filenae=app.py)

        import pickle

        #create flask app
        app= Flask(__name__)

        #Load the picke model
        model=pickle.load(open('model.pkl','rb'))

        @app.route('/predict', methods = ['POST'])
        def predict():
            json_ = request.json
            query_df =  pd.DataFrame(json_)
            prediction = model.predict(query_df)
            return jsonify({"Prediction":list(prediction)})



        if __name__ == "__main__":
            app.run(debug=True)


3) Using postman we can deploy test












Popular posts from this blog

Bagging and Boosting

  What is an Ensemble Method? The ensemble is a method used in the machine learning algorithm. In this method, multiple models or ‘weak learners’ are trained to rectify the same problem and integrated to gain desired results. Weak models combined rightly give accurate models. Bagging Bagging is an acronym for ‘Bootstrap Aggregation’ and is used to decrease the variance in the prediction model. Bagging is a parallel method that fits different, considered learners independently from each other, making it possible to train them simultaneously. Bagging generates additional data for training from the dataset. This is achieved by random sampling with replacement from the original dataset. Sampling with replacement may repeat some observations in each new training data set. Every element in Bagging is equally probable for appearing in a new dataset.  These multi datasets are used to train multiple models in parallel. The average of all the predictions from different ensemble models i...
Read more