TensorBoard and learning from Einstein | 100 Days of Code 3

In this 13‑day segment of his 100 Days of Code vlog series, Daniel Bourke narrates a highly personal, day‑by‑day chronicle of his self‑directed learning journey. He opens with a brief recap of day 11, describing his habit of single‑tasking using Pomodoro blocks, reading the Daily Stoic, and diving into Walter Isaacson’s Einstein biography. He then shifts to a technical deep‑dive on TensorBoard, explaining how the visual TensorFlow tool helps monitor neural‑network training and hyperparameter tuning. Throughout the vlog he interleaves updates on his broader learning agenda: Python study via Learn Python the Hard Way, Treehouse, DataCamp, and Udemy; research into AWS (EC2, S3, CDN) and plans to build a simple Mac timer app; ongoing podcast production (100 episodes) and Medium article writing; and personal wellness activities such as workouts, walks with his dog Seven, and time outdoors. He also recommends the newly released audiobook Barking Up the Wrong Tree and reflects on Einstein’s early career setbacks. The narrative culminates on day 13, where he outlines a “fill‑in” day of Python coding, AWS research, podcast episode 101 release, and continued use of single‑tasking and Pomodoro techniques. The vlog blends technical instruction, productivity methodology, book‑ and audiobook recommendations, and personal lifestyle habits, offering a holistic view of an independent learner’s routine.
Chapters
What's going on, y'all? Day 11. I gotta start with this hand thing. It's gonna, it's gonna be really hard once it gets to like day 95. Day 11, 100 Days of Code. What am I up to today? Well, this morning I spent, spent a couple hours in bed. I've been reading Daily Stoic as well as Einstein's biography. I just finished the sample chapter you can get on Kindle, and I ended up buying the whole book because Einstein, amazing man, amazing man. I'll share with you what I learned from the book later on. What am I up to today? Rusted a plan. It is the 22nd of May, 2017. I'm gonna write, I'm gonna work out later. 100 Days of Code, gonna write day 10. Today is day 11. Quora, I need to get a new marker. This one's out. Deep Learning course, 5 Pomodoros, 'cause I'm focusing on one task a day. So that's gonna be the number one thing today. Media article on podcasting. That's what I'm about to do now. We're uploading a YouTube video right here. The internet is going great, 47 minutes remaining. That's going to be the second edition of the vlogs I've been doing. The video I'm recording right now on this camera right now is 3. This one I'm uploading is about 14 minutes, so I don't know, I've got to work out what the length is. But I'm going to get into writing an article and then into some study, and I'll catch you later. Taking a break, enjoying the beautiful sunlight out here. Check this out, what an incredible day. This camera does nothing to highlight the beauty of what's going on here. This grass is really comfy, my shirt is camouflage. I try to get out in nature at least once per day. So I'm not behind a computer all the time. I just did some pull-ups on my pull-up bar with my faithful companion by my side. That's my dog Seven. She sometimes accompanies me on a break, but I'll be out here probably another 5 minutes or so and then I'll get back into learning. I got 3 or so Pomodoros left before my, my goal's up. And yeah, learning some cool things actually today. I'll I'll share a bit more in the next clip. And we're back. Been learning about TensorBoard, which is like a visual web application for TensorFlow. So if you're training neural networks, as you can see here, each one of these lines— well, you can't really see, this is a YouTube video. Highly recommend if you want to check it out, Hands-On TensorBoard, TensorFlow Dev Summit 2017. So it's a really recent web application. Each of these here is using different weights, different biases for each of the neural networks. So you can see which one has the best results, and it's all visual rather than sort of manually, uh, continually changing your hyperparameters every so often. If you want to find out more about hyperparameters, search up machine learning hyperparameters or deep learning hyperparameters. They're much better at, uh, explaining it than I am. But essentially, they're— if you imagine you've got a neural network, which is a bunch of algorithms, and hyperparameters would be like manually twisting the knobs, the settings on each. So if you have an oven, you want to change the temperature, you turn it to the right. And one of your hyperparameters, you could turn it to the right, and that may change the outcome of the neural network, same how it might change the temperature of your oven. So that's the main thing I've been learning about today. I've been trying— I haven't quite worked it out how to get it running properly on my computer yet, so I've been researching that. I do need a break though because my eyes are starting to hurt from reading so much. So, I figured I'm going to go for a walk soon. And I got a new audiobook that I'm excited. Barking Up the Wrong Tree: The Surprising Science Behind Why Everything You Know About Success Is Mostly Wrong. It's a brand new book. It only came out about a week ago, May 16th. Recommended by Ryan Holiday and he's someone I value valued highly in terms of book recommendations. So, you wanna check it out, audible.com, you can get your 2 first free books free. I don't have an affiliate link yet because I don't have a big enough audience, but check it out, highly recommend it. Morgen, ahoy, og god dag, it is day 12 of the 100 Days of Code series. Ah, sun, it's really bright. I just woke up, well, actually I've been up for about an hour or so now. I've been reading Einstein's biography, amazing, amazing book by Walter Isaacson. If you've never checked it out. It's a long read, it's like 800 or so pages. But I'm loving it. Getting some sun. Did you know Einstein was rejected for like a solid year from getting a job as a physicist? Assistant physicist, assistant to a professor. He applied nonstop. And I just realized I keep walking around when I'm filming. I'll try stand in one place. But yeah, he applied, it took him a couple of years after graduating from from college to actually get a job. That's Einstein. The— one of the professors he applied to several times actually never replied to him, and then 9 years later helped him out getting a Nobel Prize for his work in physics. It's crazy how the world works. What I'm up to today, single tasking again. Yesterday was all about deep learning on Udacity. That went really well. I'm really liking this, this single tasking thing. I'm going to get out of the sun. Single-tasking. So today is going to be all about Python. I'm going to head to the library and do a few hours of study there. What do I do for Python? Do the Python book, Learn Python the Hard Way, Treehouse, and then DataCamp. I'm going to check that out as well. So I'm going to do a fair bit of, a few hours Python today. And video from last week is up on YouTube. I don't know why I'm telling you this, but check it out. This clip I'm doing now will be a part of series 3. I also just wrote an article on Medium 100— I did 100 episodes of podcasts recently. You can see them all at mrdburke/podcast and the articles on my Medium, on my blog, what I learned from doing 100 days of podcasts. And episode 101 will be sometime this week. So check it out. We're back. Just finished up about, oh, 4 or 5 hours of study at the library doing Python on Treehouse and Udemy. Awesome resources. I'll put links in below if you want to check them out. But I realized what I'm doing in sort of programming, it's given me a good foundation of learning Python, doing these online courses and challenges and whatnot, but they're relatively easy to complete. If you were to say to me, okay, build me an application for this, I would struggle. So that's what I think I'll try after I finish these two courses of the fundamentals, is have an idea of some sort of app that I want to build, like just something for myself, and try to build it. And the better thing is if it's not available open source, if there's not an example, that's when I think you'll learn the most. That's what I'm— my intuition is telling me anyway, and that's what I've been reading, and that's what I've sort of— I can imagine it being so, because I— these courses and whatnot, although they're great to give you a foundation knowledge, they're sort of— they're guided, right? They got— they're like bowling with the bumper bars on. They guide the ball to the right place, so they're guiding me to be to the right outcome. So, once I finish them, I'm going to work out a project I want to work on. Something simple. Maybe a— I want a— I want an app, a simple app for my Mac that I can use to time my study sessions and then notify me when to take a break. So, maybe that's something I could work on when I'm finished, but we'll see. I'm about to go home, get something to eat, have a break, let my brain rest, go work out, and then We'll catch you in the next clip. What's good, y'all? I just finished doing some stretching, some working out. I think it's going to be it for today in terms of study. Probably not. I've got my computer here, so whenever it's in reach, I end up doing some sort of work or some sort of technology-related research. Check out the board. Didn't even update it today. Why? Because day 3 of single tasking is going really well. I don't have to do the board because I know exactly what I'm going to do. I'm just doing one task. I don't have to worry about switching my mind. So today was all about Python. I came home from the library, had some food, had a little nap, 20 minutes or so, and then went and just worked out with my brother. I just finished, so now I'm going to have some dinner. And then I'm going to help my brother out with his PowerPoint later because I'm a bit of an expert when it comes to visual presentation. Nah, I'm kidding. I've just watched Seth Godin's Prep PowerPoint course on Udemy. Phenomenal course, check it out. I recommend it to anyone who's looking to get better at presenting. I'll link it in the description for you. But that's going to be a wrap for today. Today's day 23, day 12, 100 Days of Code. I'll see you tomorrow, day 13. Unlucky for some. What's good, y'all? We're up to day 13 of 100 Days of Code. Check it out, I cleaned my whiteboard too. May 24th, 2017, day 13. I've got a few goals I want to do today. Today is more of a fill-in day, so I'm going to do a bit of Python code and then a whole bunch of other tasks I've been meaning to get to. I was supposed to spend the day with my little brother teaching him some computer science, which we're both learning through, uh, edX and the Harvard course CS50. Really good if you haven't checked it out. But he's doing something else today, so I'm just going to fill in some tasks. I've been spending the morning learning about AWS, Amazon Web Services. Boy, did I get deep. I think I spent at least 2 hours or so. I don't know, I didn't time it. 2 hours just researching how to get sites running on AWS using EC2 cloud computing, using S3 storage solutions, content delivery networks. My brain is a bit fried at the moment, so I'm gonna go get something to eat, and then I'm gonna get into some coding. Python, language of the day. I'm gonna update the Medium series today, and by the end of today I will have Podcast 101. It's coming out. Check your YouTube channel and my podcast channel for it to be somewhere there. But we'll see in the next clip. Drew up a little flowchart for a game that I'm building in Python. Because it's one of the challenges in the textbook I'm using, but getting a bit of brain block, so I'm going to go for a walk. I'm currently listening to an audiobook called Barking Up the Wrong Tree. I spoke about it the other day. Where is it? It's incredibly interesting so far. It just gives you a perspective of what— it's playing. There we go. Barking Up the Wrong Tree. It's really recent. Gives you a scientific perspective on what sort of advice is the best advice. In terms of success and whatnot. Just a whole lot of case studies and whatnot. Uh, is there actionable steps? A little bit so far. Mostly it's a bit meta, like a bit out there, just giving you two sides of the story on different things. But I don't know, I'm only about an hour in, it might get a bit better. Got a few things to do when I get back. I'm going to be recording a podcast, I'm going to upload this video to YouTube, so this will probably be the last clip. Finish off Python, I've already done two, two Pomodoros, and then fix up my Medium series. There's a little bit of an inspirational quote for myself. So that's gonna wrap up episode 3. Done a whole bunch of stuff in the last few days. I think I'm gonna start uploading maybe Wednesday and Sunday or something like that. I don't know, but this is gonna go up tonight. Thank you all so much for watching. I'm gonna get back to learning Python later after I go for a walk. Podcast 101 will be up this week— no, tonight— as well as the YouTube video to go along with it. Some of the things I've learned after doing 100 days of podcasting. Nonetheless, have an awesome day. Catch you later.
More from Daniel

Small Language Models (SLMs) Are the Future: Fine-Tuning AI That Runs on Your iPhone

Introducing Sunny - A MedGemma-Powered Skin Health Tracking App (Kaggle Competition Entry)

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 4 - Comparing the results

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 3 - Running LLMs on the RTX 4090

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 2 - Formalizing the tests

Benchmarking: NVIDIA DGX Spark & NVIDIA RTX 4090 | Part 1 - Creating a series of tests

Local Multimodal RAG on the NVIDIA DGX Spark | Part 4 - Publishing our demo to Hugging Face

Local Multimodal RAG on the NVIDIA DGX Spark | Part 3 - Putting it all together


