[Day 146] MLOps zoomcamp module 2 homework + some more prep for Microsoft x NVIDIA's hackaton

 Hello :)
Today is Day 146!


A quick summary of today:
  • did 2nd homework from MLOps zoomcamp on experiment tracking
  • covered some material on devops on Azure


MLOps zoomcamp Module 2 homework on experiment tracking

The homework link is here. The homework followed the below outline:

  1. Install mlflow
  2. Preprocess NYC taxi data
  3. Train a model with autolog
  4. Launch a tracking server locally
  5. Tune model hyperparams
  6. Promote the best model

My code for it is on my github.

The used model was a random forest regressor using rmse as the main evaluation metric. The homework was more focused on running mlflow commands rather than the preprocessing/training (which were done in Module 1 of the course). 

Below is a graphical comparison of some of the models I ran


I had to set up logging and mlflow db for a sample run

Run hyperparam optimization 

And finally choosing and registering the best model

For a bit I faced an issue with the params passed to the model because they were missing (initially not logged in the hyperparam opt code), and then I kept registering the *worst* performing model because i had [0] at the end of that best run, but when I put [-1] it was all good. Below are the top 5 and then best among the 5 models is registerered. 


As for the next module - module 3, I saw we will use Mage to learn Orchestration and ML Pipelines and saw that the youtube videos were uploaded today. 

2nd prep course for Microsoft x NVIDIA hackaton in Microsoft's office in Seoul

It was a bit long so I covered just the bits I thought were new, and below are some notes I took along the way.

Azure Boards: tools for agile planning, work item tracking, visualization, and reporting

Azure Pipelines: a CI/CD platform that is language, platform, and cloud-agnostic, supporting containers and Kubernetes

Azure Repos: cloud-hosted private Git repositories with two types of version control: distributed (Git) and centralized (Team Foundation Version Control, TFVC). Features include

  • Free private Git repositories
  • Support for any Git client
  • Web hooks and API integration
  • Semantic code search
  • Git code reviews with threaded discussion
  • Continuous integration for each change
  • Branch policies to ensure code quality
  • Integration with favorite tools and editors

Azure Artifacts: Integrated package management supporting Maven, npm, Python, and NuGet packages from public or private sources

Azure Test Plans: An integrated solution for planned and exploratory testing

Workflows: Define automation processes, detailing trigger events and jobs to run. Workflows are written in YAML and located within the GitHub repository at .github/workflows. Example:

GitHub runners: compute resources that execute GitHub Actions workflows, performing build, test, and deployment tasks directly within GitHub repositories

Continuous integration with actions example

  • On: Specifies what will occur when code is pushed
  • Jobs: There's a single job called build
  • Strategy: It's being used to specify the Node.js version
  • Steps: Are doing a checkout of the code and setting up dotnet
  • Run: Is building the code

Next, Azure Pipelines is a robust service for creating cross-platform CI/CD workflows, integrating with your preferred Git provider and deploying to major cloud services. Key features

  • End-to-end flow of value: Focus on delivering continuous value from concept to customer, avoiding siloed processes
  • Automation and reliability: Establish a repeatable, reliable process for software delivery
  • Quality verification: Each pipeline stage verifies feature quality to prevent user-facing errors
  • Team feedback and visibility: Ensure all team members have feedback and visibility into the delivery process
  • Frequent small changes: Promote frequent, smaller changes for better flow and optimization
  • Continuous improvement: Continuously monitor and resolve obstacles to improve pipeline efficiency

Pipeline stages:

  • Build automation and Continuous Integration: Automatic building and integrating of changes
  • Test automation: Automated testing for functionality and quality
  • Deployment automation: Automated deployment to various environments

Orchestration:

  • Release and pipeline orchestration: Manage and control the entire pipeline, providing top-level insights
  • Value stream mapping: Identify and improve inefficiencies
  • Infrastructure efficiency: Efficient infrastructure is crucial for an effective pipeline


Key terms in Azure pipelines


  • Agent: software that runs a build or deployment job
  • Artifact: collection of files or packages published by a build, used for distribution or deployment
  • Build: one execution of a pipeline, collecting logs and test results
  • Continuous Delivery (CD): process of building, testing, and deploying code to various stages, ensuring quality through multiple stages and constant monitoring
  • Continuous Integration (CI): practice of automating the testing and building of code to catch issues early. CI systems produce artifacts used in CD pipelines for automatic deployments
  • Deployment target: destination for application deployment, such as a virtual machine, container, or web app
  • Job: a set of steps run on an agent within a build, representing an execution boundary
  • Pipeline: script defining the CI/CD process, composed of tasks for testing, building, and deploying
  • Release: execution of a release pipeline, including deployments to multiple stage
  • Stage: primary divisions in a pipeline, such as 'build,' 'test', and 'deploy'
  • Task: building block of a pipeline, performing specific jobs like building, testing, or deploying
  • Trigger: configuration that specifies when the pipeline should run, such as on code push or schedule


That is all for today!

See you tomorrow :) 

Popular posts from this blog

[Day 198] Transactions Data Streaming Pipeline Porject [v1 completed]

Image
 Hello :) Today is Day 198! A quick summary of today: data streaming pipeline project [v1 done] Here is a link to the project's repo. Well ... I did not know I can do it in a day (~14 hours) after yesterday's issues but here we are. Turns out in order to insert the full (~70 variables with nested/list structure), I need the proper pyspark schema. And yesterday I did not have that and that is why when I was reading data in the kafka producer I was getting NULL in the columns - my schema was wrong. Well today I not only fixed the schema for the 4 variables I had yesterday, but included *all* the variables that come from the Stripe API ~ 70 (for completeness).  When I run docker-compose, the data streams and is input into the postgres db (and is still running). Unfortunately, the free Stripe API for creating realistic transactions has a limit of 25, so every 3 seconds, 25 new transactions are sent to the db. It has been running half the day (since I got that set up) and as I am wr
Read more

[Day 107] Transforming natural language to charts

Image
 Hello :) Today is Day 107! A quick summary of today: created text2chart - transforming natural language to charts [ github ] [ webapp ] I am super sick, but in the around 3 hours of my fever not being high, I managed to study a bit. I was again thinking about creating some nlp project but was stuck on the idea of how to host a model. After looking around I found this article where they create a similar app called chat2vis  and they use codellama. I know codellama is an open source model that I can get from huggingface and I wondered, how did they host it. Turns out - they ask the user to input a huggingface api key, which then allows for querying and generating content. (AWESOME! this is applicable to some of the other PDF chat apps that I have, but I will look to make an app of them later).  My fever eased a bit, and I sat on my chair to do some work ^^ Here is a breakdown of the app. I used streamlit. Also, 2 default datasets during dev: Financial Statements of Major Companies(2009-
Read more

[Day 54] I became a backprop ninja! (woohoo)

Image
 Hello :) Today is Day 54! A quick summary of today: Read through cs231n's neural network lectures [1] [2] [3] Spent 8 hours going over Andrej Karpathy's backprop ninja tutorial After yesterday's confidence boost in doing backprop, I came in today with the intention of becoming of, as Andrej Karpathy calls it, a backprop ninja.  The cs231n NN lectures were the usual about how NN became to be, param initialization, activation functions, loss functions, regularization and optimizers. Nothing out of the ordinary, but still was good to reassure myself of that knowledge. But that was definitely not the main even of today. As I mentioned, I wanted to become (or at least attempt to become) a backprop ninja today. I have watched that particular tutorial twice, the 1st time I had no idea what is going on, the 2nd time I could follow, and now it was the 3rd time - and now, I wanted to participate and actually see the why and the how of the things we have to do, when we do manual back
Read more