100/100 Days of ML Challenge
It’s been a long journey, but I’ve finally finished the 100 Days of ML series. I’ve learned more than I imagined through consistent and persistent effort. When I started the series, India was in the middle of the second wave COVID-19 pandemic and by the time I ended, I finished my summer internship at IISc and my 4th-semester finals concurrently.
My previous article “50/100 Days of ML Challenge — Halfway Evaluation” talked about how to set up a machine learning roadmap. In this article, I’m going to talk more about my experiences rather than providing a guideline.
Importance of Planning
Meticulous structuring often gets overlooked during the initial phases of creating a project. One thing I’ve learned from streaming for 100 days is the value of creating hierarchical structures while planning a project. When trying to establish a plan for a project I would recommend breaking it up into three distinct hierarchical groups — inference, collective, and immediate.
The inference and collective goals are both defined during the initial stages of the project. The inference is a concise descriptive paragraph of what the project entails. This paragraph would act as a strong reference point to build the collective goals.
The collective goals are a set of overarching tasks which encompasses the primary objectives for successfully completing the project. Although these tasks would be constructed at the initial stages of the project, It would be recommended to adapt and review these goals every quarter of the project timeline.
The final sector defines the immediate goals. Although these goals fall at the bottom of the hierarchy they are the most crucial to set up. The process of building the inference and collective goals is to ultimately provide a strong foundation in order to seamlessly establish immediate goals on a day-to-day basis. These high-frequency immediate goals are meant to describe the objectives within a short duration. Ultimately by completing a sequence of short goals you would be able to check off an overarching task from the collective goals pile.
For example, the GitHub README.md file for this project consists of “Description”, “Challenge Goals” and “Daily Goals” which are analogous to the inference, collective goals, and immediate goals. Over time I realized I had to pivot and make changes to my challenge goals. I also didn’t complete certain challenge goals after realizing that It wasn’t in my best interest. These are all normal spontaneous events. Never try to follow a set of goals rigidly, always try to give yourself enough freedom to go with the flow. I believe that this hierarchical structure for planning allows for that.
Creating Healthy Habits
Streaming on youtube every day helped me establish an incredibly powerful habit of performing deep work for a certain period of time. For a significant portion of the challenge, I would wake up at 5:30 AM and stream for 30 min till 6 AM. I would then continue to be productive till 11 AM. Before I knew it I was getting five and a half hours of work done before most people even woke up.
Start by establishing one rigid healthy habit and soon enough more will follow. I suggest reading “Deep Work” by Cal Newport and “The Power of Habit” by Charles Duhigg. Reading these books could be your initial habit :P
Being Persistent
One of the hardest challenges of performing a strenuous task consistently is the ability to be diligent. Some of the ways you could ease the burden are by spending time organizing yourself and making sure that your “plan” is solid. These preparations will ensure that you can get straight into deep focus instead of having to worry about what to do.
I hope this article has helped you in some way or another. If you would like to know more about the challenge I recommend checking out my GitHub repository or follow my progress on YouTube. If you would like to know more about me, consider visiting my website harshithmohankumar.com