Deploy Machine Learning model using Streamlit

   March 6, 2022
Deploy Machine Learning model using Streamlit

Machine Learning model is software that can learn and respond through experiences. Machine Learning has a wide range of applications in various fields of technology and science. Many top companies give preference to machine learning as one of the most important domains. For example, Amazon, Netflix, Facebook, Youtube all these companies use Machine Learning algorithms to improve customer experience. Since Machine Learning is a much established and more blooming field, it is a great career option one may choose. So, in this article, we will learn how to Deploy Machine Learning model using Streamlit.
The ultimate goal of the Machine Learning model is to make use of provided data and make better predictions. There are some steps to build a machine learning model. https://copyassignment.com/8-steps-to-build-a-machine-learning-model/. One of the most important and final steps in building a Machine Learning project is Model deployment. There are many frameworks available for deploying the Machine learning model on the web. Some of the most used Python frameworks are Django and Flask. But these frameworks require a little knowledge of languages such as HTML, CSS, and JavaScript. So, a new framework known as Streamlit was introduced to deploy the Machine Learning model without the need of having the knowledge of Front End Languages. It is quite easy to deploy using Streamlit.

What is Streamlit?

Streamlit is an open-source and user-friendly framework designed to create and deploy Machine Learning models. It is used to deploy the Machine Learning models as web apps in a minimal amount of time. Streamlit is much simpler and requires a minimal amount of code compared to other frameworks. Since Streamlit is a new framework lot of new features are being added in the subsequent releases.

Features of Streamlit

  • The web app created using Streamlit automatically gets updated when the changes are made in source file.
  • Streamlit framework is compatible with most of the Python libraries such as Numpy, Pandas, PyTorch, Matplotlib, Keras, etc.,
  • The code to deploy our model is the same way as we write Python code.
  • No need to worry about the front end development.
  • It is extremely easy and simple to learn.
  • The development and deployment time is super fast.
  • Since widgets are treated as variables it is quite easy to implement.
assignment advertisement

How to install Streamlit?

Before installing Streamlit, make sure you have installed Python in your system.

To install Streamlit run the following command in the command prompt or terminal.

pip install streamlit

Open a python file and import the Streamlit framework use the following command

import streamlit as st

To deploy and run the web app on the local web run the following command.

streamlit run filename.py

The web app runs on port 8501 by default. You can check the output in the browser. After making any changes to the source code, rerun the app using the option available over the right top corner. Or use the option always rerun to automatically update the changes to the web app.

Important functions in Streamlit

Streamlit has many features and functions to deploy the content on the web. Some of the important and most used functions are discussed below.

Title

The title widget is used to specify the title for the page.

# Importing Streamlit
import streamlit as st
# Title widget
st.title(“Deployed with Streamlit”)

Output:

Deploy Machine Learning model using Streamlit

Write

Write function accepts data formats, such as text, numbers, data frames, styled data frames, and assorted objects.

import streamlit as st
import pandas as pd
# Write widget
st.write (“This is a Web App”)
st.write (1234)
st.write (pd.DataFrame({
‘first col’: [1, 2, 3, 4],
‘second col’: [1, 4, 6, 8],
}))

Output:

write output in table form to deploy

Text elements

Some of the text elements include the header, Subheader, caption, text, and latex.

import streamlit as st
# Text widgets
st.header (“This is a header”)
st.subheader (“This is a subheader”)
st.caption (“This is a caption”)
st.text (“This is a text”)

Output:

Text output

Code widget is used to specify the code on the page.

import streamlit as st
code = ”’def function():
print (“This is a web app!”)”’
# Code widget
st.code (code, language=’python’)

Output:

Code output

Text color

import streamlit as st
st.success (“Success”)
st.info (“Info”)
st.warning (“Warning”)
st.error (“Error”)

Output:

text output

Check box and Radio button

import streamlit as st
st.text (‘Check Box’)
#checkbox
st.checkbox (‘Option 1’)
st.checkbox (‘Option 2’)
#radio button
st.radio (“Radio button”,(‘Option1′,’Option2’))

Output:

Deploy Machine Learning model using Streamlit

Selection Box

Any one option can be selected from the selection box. In Selection box first argument takes the title of the selection box second argument takes the options.

import streamlit as st
#Selection box
st.selectbox (“Selection box “, [‘Option 1’, ‘Option 2’, ‘Option 3’])

Output:

selection box

Text Input and Button

import streamlit as st
st.text_input (“Text Input”)
st.button (“Button”)

Output

Text Input and Button ouput

These are the common widgets in the streamlit framework. Let us now implement a Machine Learning project and deploy using the Streamlit framework. We are going to build the machine learning model for Heart disease prediction.

Heart Disease prediction with Machine Learning and deploy using streamlit

Machine Learning is very useful in the medical field to detect many diseases in the early stage. Heart disease prediction is one such Machine Learning model which helps to detect heart disease in humans.

The workflow to build the end to end Machine Learning Project to predict heart disease is as follows

  1. Collection of data
  2. Data Preprocessing
  3. Splitting the data
  4. Training the model
  5. Predicting the outcome
  6. Deploying the model

Data collection

The very first step is to choose the dataset for our model. We can get a lot of different datasets from Kaggle. You just need to sign in to Kaggle and search for any dataset you need for the project. The Heart disease dataset required for our model can be downloaded with the following link.

Heart Disease Dataset

The Heart Disease dataset contains about 14 columns and the details available in the dataset includes,

  1. age – age in years
  2. sex – (1 = male; 0 = female)
  3. cp – chest pain type (4 values)
  4. trestbps – resting blood pressure (in mm Hg on admission to the hospital)
  5. chol – serum cholesterol in mg/dl
  6. fbs – fasting blood sugar > 120 mg/dl (1 = true; 0 = false)
  7. restecg – resting electrocardiographic results (values 0,1,2)
  8. thalach – maximum heart rate achieved
  9. exang – exercise induced angina (1 = yes; 0 = no)
  10. oldpeak – ST depression induced by exercise relative to rest
  11. slope – the slope of the peak exercise ST segment
  12. ca – number of major vessels (0-3) colored by fluoroscopy
  13. thal – (0 = normal; 1 = fixed defect; 2 = reversible defect)
  14. target – 0 or 1

Importing the required libraries

After downloading the dataset, import the necessary libraries required for building the model before you Deploy Machine Learning model using Streamlit

#Importing the libraries
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

Load the data using the read_csv method in the pandas library

#Loading the data
data = pd.read_csv (‘data.csv’)

The head() method in the pandas library is used to print the rows up to the limit we specify. The default number of rows is five.

#Print the first 5 rows of the dataset
data.head()

Output:

First five rows of the dataset

#To get the number of rows and columns in the dataset
data.shape

Output:
(1025, 14)

We can see that there are 1025 rows and 14 columns in the dataset

#Getting information about the data
data.info()

Output:

Information about the data

From the information we collected in the data, we can see that there is no null or missing data in the dataset. And, it is clear that Target is the outcome variable. So let us explore more details about that column.

#Checking the distribution of Target Variable
data[‘target’].value_counts()

Output:
1 526
0 499
Name: target, dtype: int64

In the target column, the value 1 means the person is having Defective Heart and 0 means the person is having Healthy Heart

Splitting the data

The next step in the building of the Machine learning model is splitting the data into training and testing sets. The training and testing data should be split in a ratio of 3:1 for better prediction results.

X = data.drop (columns=’target’, axis=1)
Y = data [‘target’]

print(X)

Output:

Output of X

print(Y)

Output:

Output of Y

Splitting the Data into Training data & Test Data

X_train, X_test, Y_train, Y_test = train_test_split (X, Y, test_size=0.2, stratify=Y, random_state=2)
print (X.shape, X_train.shape, X_test.shape)

Output:
(1025, 13) (820, 13) (205, 13)

Training the model

The next step is to build and train our model. We are going to use a Logistic regression algorithm to build our model. Logistic Regression is one of the popular supervised Machine Learning algorithms. It is used to predict the outcome of categorical data. Classification problems like Heart disease prediction can be implemented using a Logistic Regression algorithm.

model = LogisticRegression()
# Training the LogisticRegression model with the Training data
model.fit (X_train.values, Y_train)

Model Evaluation

After building the model, the model has to predict output with test data.

# Predicting the output with test data
y_pred=model.predict (X_test.values)
print (y_pred)

Output:

Deploy Machine Learning model using Streamlit

After the prediction of outcome with test data, we can calculate the accuracy score of the prediction results by the model.

#Calculating the accuracy of the predicted outcome
print (accuracy_score (Y_test,y_pred))

Output:
0.8048780487804879

Building a Predictive System

input_data = (62, 0, 0, 140, 268, 0, 0, 160, 0, 3.6, 0, 2, 2)
# Change the input data to a numpy array
numpy_data= np.asarray (input_data)
# reshape the numpy array as we are predicting for only on instance
input_reshaped = numpy_data.reshape (1,-1)
prediction = model.predict (input_reshaped)
if (prediction[0]== 0):  
print (‘The Person does not have a Heart Disease’)
else:  
print (‘The Person has Heart Disease’)

Output:
The Person does not have a Heart Disease

#Saving the trained model
import pickle
filename = ‘trained_model.sav’
#dump=save your trained model
pickle.dump (model,open (filename,’wb’))
#loading the saved model
loaded_model = pickle.load (open (‘trained_model.sav’,’rb’))

Once you run this code a new file named trained_model.sav will be saved the project folder.

Python Code to Deploy Machine Learning model using Streamlit

Open a new Python file and execute the following code.

App.py

#Importing the libraries
import numpy as np
import pickle
import streamlit as st

#loading the saved model
loaded_model = pickle.load (open (“C:/Downloads/trained_model.sav”, ‘rb’))

#Creating a function for Prediction
def heartdisease_prediction (input_data):
# changing the input data to a numpy array
numpy_data= np.asarray (input_data)
#Reshaping the numpy array as we are predicting for only on instance
input_reshaped = numpy_data.reshape (1,-1)
prediction = loaded_model.predict (input_reshaped)
if (prediction[0] == 0):
st.success (‘The person does not have heart disease’)
else:
st.warning (‘The person have heart disease’)

#Adding title to the page
st.title (‘Heart disease prediction Web App’)

#Getting the input data from the user
age = st.text_input (‘Age in Years’)
sex = st.text_input (‘Sex : 1 – male, 0 – female’)
cp = st.text_input (‘Chest pain type’)
trestbps = st.text_input (‘Resting blood pressure in mm Hg’)
chol = st.text_input (‘Serum cholesterol in mg/dl’)
fbs = st.text_input (‘Fasting blood sugar > 120 mg/dl : 1 – true, 0 – false’)
restecg = st.text_input (‘Resting electrocardiographic results’)
thalach = st.text_input (‘Maximum heart rate achieved’)
exang = st.text_input (‘Exercise induced angina: 1 – yes, 0 – no’)
oldpeak = st.text_input (‘ST depression’)
slope = st.text_input (‘Slope’)
ca = st.text_input (‘Number of major vessels (0-3)’)
thal = st.text_input (‘Thal’)

# code for Prediction
diagnosis = ‘ ‘
# creating a button for Prediction
if st.button (‘Heart Disease Test Result’):
diagnosis=heartdisease_prediction ([age,sex,cp,trestbps,chol,fbs,restecg,thalach,
exang,oldpeak,slope,ca,that])

Save the file after pasting the code. And then to deploy using streamlit go to command prompt run the following command.

streamlit run App.py
(or)
streamlit run filename.py

After running the command the web app will open in the localhost webserver. Otherwise, go to your browser and type localhost:8501. The following output will be shown.

Output:

Deploy Machine Learning model using Streamlit
deploy using streamlit

Sample Input data for person does not have a heart disease is {52, 1, 0, 125, 212, 0, 1, 168, 0, 1, 2, 2, 3}. These data as input will generate the following output in the web app.

deploy using streamlit

Sample input data for a person who have a heart disease is {58, 0, 0, 100, 248, 0, 0, 122, 0, 1, 1, 0, 2}.
These data as input will generate the following output in the web app.

deploy ml using streamlit

Conclusion

Thus in this article, we learned how to Deploy Machine Learning model using Streamlit. Although Streamlit has many advantages over the other frameworks, lot more features are under development. So if you are getting started in Machine Learning then this framework will be a perfect start to deploy your machine learning model on the web.

Also Read:

Share:

Author: Ayush Purawr