https://github.com/machinelearningzuu/deeplearning.ai-machine-learning-engineering-for-production-mlops

https://github.com/machinelearningzuu/deeplearning.ai-machine-learning-engineering-for-production-mlops

Last synced: 4 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/machinelearningzuu/deeplearning.ai-machine-learning-engineering-for-production-mlops
• Owner: machinelearningzuu
• Created: 2021-09-11T14:55:09.000Z (almost 5 years ago)
• Default Branch: main
• Last Pushed: 2021-09-12T04:19:45.000Z (almost 5 years ago)
• Last Synced: 2025-02-04T17:17:42.809Z (over 1 year ago)
• Size: 1.73 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# DeepLearning.AI

# Machine Learning Engineering for Production [MLOps] Specialization

### This Repository contains Course materials & assignments related to Coursera Machine Learning Engineering for Production Specialization. MLOps means Machine Learning operations. In simple terms using MLOps we can describe the Industrial Machine Learning Pipeline for large scale projects.

## MLOps Basic Workflow

![github](https://github.com/1zuu/1zuu-DeepLearning.AI-Machine-Learning-Engineering-for-Production-MLOps/blob/main/mlops.jpg)

# Course 1 - Introduction to Machine Learning in Production

### In this course it gives basic intuition of Machine Learning Project life cycle. 

## Machine Learning Project Life Cycle

![github](https://github.com/1zuu/DeepLearning.AI-Machine-Learning-Engineering-for-Production-MLOps/blob/main/Course%201%20-%20Introduction%20to%20Machine%20Learning%20in%20Production/Week%201/notes/ML%20Project%20Life%20Cycle.PNG)

### 1. Scoping Stage

![github](https://github.com/1zuu/DeepLearning.AI-Machine-Learning-Engineering-for-Production-MLOps/blob/main/Course%201%20-%20Introduction%20to%20Machine%20Learning%20in%20Production/Week%201/notes/scoping.PNG)

#### Here we define the main objectives of the project. What are the key metrics should focus for evaluation? What kind of trade-offs (speed vs accuracy) we have to face ? are the main question we should ask in this stage. This is such key stage of the entire life cycle.

### 2. data Stage

![github](https://github.com/1zuu/DeepLearning.AI-Machine-Learning-Engineering-for-Production-MLOps/blob/main/Course%201%20-%20Introduction%20to%20Machine%20Learning%20in%20Production/Week%201/notes/define%20data.PNG)

### 3. modeling Stage

![github](https://github.com/1zuu/DeepLearning.AI-Machine-Learning-Engineering-for-Production-MLOps/blob/main/Course%201%20-%20Introduction%20to%20Machine%20Learning%20in%20Production/Week%201/notes/modeling.PNG)

### 4. deployment Stage

![github](https://github.com/1zuu/DeepLearning.AI-Machine-Learning-Engineering-for-Production-MLOps/blob/main/Course%201%20-%20Introduction%20to%20Machine%20Learning%20in%20Production/Week%201/notes/deployment.PNG)

#### One of the Most Challenge in Deployment Stage is DATA DRIFT & CONCEPT DRIFT. DATA DRIFT means that the distribution of data may changes from training phase to inference based on variety of software, hardware & other issues. CONCEPT DRIFT means input output mapping of a supervised learning task changes. As as Example, Consider speech recognition system. We only trained the system for data collected from Teenagers. But due to pandemic old age people use the system for variety of reasons.