Machine Learning

Flask is web application framework in python. It's built on the Werkzeug WSGI toolkit and Jinja2 template engine.  Web server gateway interface (WSGI) helps in requests, response object and other utility functions. Jinja2 : templating engine helps in rendering dynamic web pages.  Pip install Flask  ML model deployment using flask Step 1 : Pickling the ML model  import pickle  pickle.dump(model1, open('myfirst_model.pkl','wb')) Pickle module help in serializing ML model and save to a file.  Later we can load model from file and deserialize to make predictions. model = pickle.load(open('myfirst_model.pkl','rb')) 

Machine learning is part of Artificial Intelligence (AI) that help computer to learn without explicit programming.  - learn from data and predict pattern. There are 2 type of Machine learning category mainly. 1. Supervised learning : need some existing sample data with output level for training. Then it can predict bases on training. Example Classification and Regression. 2. Unsupervised learning : there is no output level for training. It get data and try group data based on similarities.  Steps involved in creating ML models. 1. Train the Model 2. Package model 3. Validate model 4. Deploy model 5. Monitor performance Some tools to help with ML development. 1. Apache Airflow : schedule, monitoring workflow or data pipeline. 2. Kubeflow : Orchestrate complicated workflow running on kubernetes. 3. MLflow : tracking ML experience, deployment of the model. 4. Flask : framework to deploy ML models. 5. Seldon : deployment in kubernetes at a massive scale. 

 While executing PowerShell script if you get  + FullyQualifiedErrorId : UnauthorizedAccess error.  Please follow below steps. 1. Execute >Get-ExecutionPolicy -Scope CurrentUser  If this returns Undefined then we have to give access to current user to execute PowerShell command.  2. Execute >Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser 3. Now you should be able to run Powershell Command.  4. Once you have done with work then again revoke access as best practice.  5. Execute >Set-ExecutionPolicy -ExecutionPolicy Undefined -Scope CurrentUser Activate Virtual Environment using below command. 

Evolution of computer is beyond comparison. Something extra ordinary, It hard to believe that we will achieve so much in this field in this small time interval. In this blog we are going to discuss about some of those most important event in history of computes.  1. 1950 : Vacuum Tubes  takes lot of energy.  2. Later 1950 : Transistors  Same functionality as vacuum tubes but takes less space and energy using semiconductor. Transistor radio.  3. 1960 : The Integrated Circuits (IC) millions of tiny transistors in small board.  4. 1970 : The Personal Computer  Datapoint 2200 first personal computer.  5. 1975 : Altair 8800. New Basic language. Microsoft Bill Gates and Paul Allen. 6. 1976 : Steve Jobs and Steve Wozniak started Apple. Single-circuit board.  7. 1981 : IBM Model 5150 with 4.77MHz Intel processor and Microsoft's MS-DOS operating system.  8. 1983 : Apple's Lisa with graphic user interface (GUI). Mouse 9. 1984 : Apple Macintosh and...

 In this we are going to explain about Type I and Type II error like false positive and false negative. False positive False - Outcome whether is was true or False Positive - It's action or indication by device or process.  False Positive : Device say it's there which is in actual false mean in actual it was not there . False alarm. False negative : kind of device is not working. negative means device say it's not there. Which is false statement. so in actual it's there but device not able to detect. Broken Alarm.  Example: There is a Alien detector machine. Which indicate if there is any alien nearby.  False positive : positive means detector say there is an Alien which is false. In actual there is no Alien. This situation is fine as we might check manually for Alien or get alert. Even though it's false alarm not getting any harm.  False Negative : Negative means detector say there is no Alien which is false. In actual there is an Alien. This is not good be...

 Visualization help us to understand data in better way. This is specially important when we have lot of data say million or billions of records. Just looking at data we might not get the relationship or inside trend of data.  Today, we are going to learn about line chart using matplotlib library.  from matplotlib import pyplot as plt years = [2009, 2011, 2012, 2012.5, 2013, 2013.5, 2014, 2014.5, 2015, 2015.5, 2016] scores = [76.6, 80, 88, 88, 87, 88, 87, 87, 85, 87, 86] plt.plot(years, scores, color='blue', marker='X', linestyle='dashdot') plt.title("Performance") plt.ylabel("Score %") plt.show() With example we learned how to create line chart. Now we will learn where to use to line chart.  Trends : If we wanted to show trend like search engine preferred by people over time or stock time change over time.