OK, now we're talking, two liner and you get a window showing bounding box object detection from your webcam feed!The accuracy is crap for anything but people. But still. Well done. Tested on Ubuntu 22.10, P51.
python -m pip install -U yolov5==7.0.9
yolov5 detect --source 070,000 28x28 grayscale (1 byte per pixel) images of hand-written digits 0-9, i.e. 10 categories. 60k are considered training data, 10k are considered for test data.
This is THE "OG" computer vision dataset.
Playing with it is the de-facto computer vision hello world.
It was on this dataset that Yann LeCun made great progress with the LeNet model. Running LeNet on MNIST has to be the most classic computer vision thing ever. See e.g. activatedgeek/LeNet-5 for a minimal and modern PyTorch educational implementation.
But it is important to note that as of the 2010's, the benchmark had become too easy for many applications. It is perhaps fair to say that the next big dataset revolution of the same importance was with ImageNet.
The dataset could be downloaded from yann.lecun.com/exdb/mnist/ but as of March 2025 it was down and seems to have broken from time to time randomly, so Wayback Machine to the rescue:but doing so is kind of pointless as both files use some crazy single-file custom binary format to store all images and labels. OMG!
wget \
https://web.archive.org/web/20120828222752/http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz \
https://web.archive.org/web/20120828182504/http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz \
https://web.archive.org/web/20240323235739/http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz \
https://web.archive.org/web/20240328174015/http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz
OK-ish data explorer: knowyourdata-tfds.withgoogle.com/#tab=STATS&dataset=mnist
If you are a pussy and work a soul crushing job, this is one way to lie to yourself that your life is still worth living: do one cool thing every day.
Find a time in which your mind hasn't yet been destroyed by useless work, usually in the morning before work, and do one thing you actually like in life.
Work a little less well for you boss, and a little better for yourself. Ross Ulbricht:Selling drugs online is not advisable however.
Even better, try to reach an official agreement with your employer to work 20% less than the standard work week. For example, you could work one day less every week, and do whatever you want on that day. It is not possible to push your passion to weekends, because your brain is too tired. "You keep all non-company-related IP you develop on that time" is a key clause obviously.
On a related note, good employers must allow employees to do whichever the fuck "crazy projects", "needed refactorings or other efficiency gains" and "learn things deeply" at least 20% of their time if employees want that: en.wikipedia.org/wiki/20%25_Project. Employees must choose if they want to do it one day a week or two hours per day. One day per month initiatives are bullshit. Another related name: genius hour.
Highly relevant on this topic: Video "What Predicts Academic Ability? by Jordan B Peterson (2017)".
I did it for me, Skyler
. Source. Pursuing a dream part time can make you feel afraid and tired. But at least, you will feel alive.Maybe you will be fired, but long term, having tried, or even succeeded your dream, or a one of its side effects, will be infinitely more satisfying.
The same goes for school, and maybe even more so because your parents can still support you there. Some Gods who actually followed this advice and didn't end up living under a bridge:
- George M. Church "[We] hope that whatever problems... contributed to your lack of success... at Duke will not keep you from a successful pursuit of a productive career." Lol, as of 2019 the dude is the most famous biotechnologist in the world, those "problems" certainly didn't keep him back.
- Freeman Dyson proved the equivalence of the three existing versions of quantum electrodynamics theories that were around at his time, and he has always been proud of not having a PhD!
- Ramanujan, from Wikipedia:
He received a scholarship to study at Government Arts College, Kumbakonam, but was so intent on mathematics that he could not focus on any other subjects and failed most of them, losing his scholarship in the process.
- Person that Ciro met personally and shall remain anonymous for now for his privacy: once Ciro was at a bar with work colleagues casually, it was cramped, and an older dude sat next to his group.The dude then started a conversation with Ciro, and soon he explained that he was a mathematician and software engineer.As a Mathematician, he had contributed to the classification of finite simple groups, and had a short Wiki page because of that.He never did a PhD, and said that academia was a waste of time, and that you can get as much done by working part time a decent job and doing your research part time, since you skip all the bullshit of academia like this.Yet, he was still invited by collaborating professors to give classes on his research subject in one of the most prestigious universities in the world. Students would call him Doctor X., and he would correct them: Mister X.As a software engineer, he had done a lot of hardcore assembly level optimizations for x86 for some mathematical libraries related to his mathematics interests. He started talking microarchitecture with Ciro's colleagues.
TODO where to find it: www.kaggle.com/general/50987
Subset of ImageNet. About 167.62 GB in size according to www.kaggle.com/competitions/imagenet-object-localization-challenge/data.
Contains 1,281,167 images and exactly 1k categories which is why this dataset is also known as ImageNet1k: datascience.stackexchange.com/questions/47458/what-is-the-difference-between-imagenet-and-imagenet1k-how-to-download-it
www.kaggle.com/competitions/imagenet-object-localization-challenge/overview clarifies a bit further how the categories are inter-related according to WordNet relationships:
The 1000 object categories contain both internal nodes and leaf nodes of ImageNet, but do not overlap with each other.
image-net.org/challenges/LSVRC/2012/browse-synsets.php lists all 1k labels with their WordNet IDs.There is a bug on that page however towards the middle:and there is one missing label if we ignore that dummy
n02119789: kit fox, Vulpes macrotis
n02100735: English setter
n02096294: Australian terriern03255030: dumbbell
href="ht:
n02102040: English springer, English springer spanielhref= line. A thinkg of beauty!Also the lines are not sorted by synset, if we do then the first three lines are:
n01440764: tench, Tinca tinca
n01443537: goldfish, Carassius auratus
n01484850: great white shark, white shark, man-eater, man-eating shark, Carcharodon carchariasgist.github.com/aaronpolhamus/964a4411c0906315deb9f4a3723aac57 has lines of type:therefore numbered on the exact same order as image-net.org/challenges/LSVRC/2012/browse-synsets.php
n02119789 1 kit_fox
n02100735 2 English_setter
n02110185 3 Siberian_huskygist.github.com/yrevar/942d3a0ac09ec9e5eb3a lists all 1k labels as a plaintext file with their benchmark IDs.therefore numbered on sorted order of image-net.org/challenges/LSVRC/2012/browse-synsets.php
{0: 'tench, Tinca tinca',
1: 'goldfish, Carassius auratus',
2: 'great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias',The official line numbering in-benchmark-data can be seen at
LOC_synset_mapping.txt, e.g. www.kaggle.com/competitions/imagenet-object-localization-challenge/data?select=LOC_synset_mapping.txtn01440764 tench, Tinca tinca
n01443537 goldfish, Carassius auratus
n01484850 great white shark, white shark, man-eater, man-eating shark, Carcharodon carchariasThe official page: www.image-net.org/challenges/LSVRC/index.php points to a download link on Kaggle: www.kaggle.com/competitions/imagenet-object-localization-challenge/data Kaggle says that the size is 167.62 GB!
To download from Kaggle, create an API token on kaggle.com, which downloads a The download speed is wildly server/limited and take A LOT of hours. Also, the tool does not seem able to pick up where you stopped last time.
kaggle.json file then:mkdir -p ~/.kaggle
mv ~/down/kaggle.json ~/.kaggle
python3 -m pip install kaggle
kaggle competitions download -c imagenet-object-localization-challengeAnother download location appears to be: huggingface.co/datasets/imagenet-1k on Hugging Face, but you have to login due to their license terms. Once you login you have a very basic data explorer available: huggingface.co/datasets/imagenet-1k/viewer/default/train.
From cocodataset.org/:
So they have relatively few object labels, but their focus seems to be putting a bunch of objects on the same image. E.g. they have 13 cat plus pizza photos. Searching for such weird combinations is kind of fun.
Their official dataset explorer is actually good: cocodataset.org/#explore
Also, images have captions describing the relation between objects:Epic.
This dataset is kind of cool.
As of v7:
- ~9M images
- 600 object classes
- bounding boxes
- visual relatoinships are really hard: storage.googleapis.com/openimages/web/factsfigures_v7.html#visual-relationships e.g. "person kicking ball": storage.googleapis.com/openimages/web/visualizer/index.html?type=relationships&set=train&c=kick
- google.github.io/localized-narratives/ localized narratives is ludicrous, you can actually hear the (Indian women mostly) annotators describing the image while hovering their mouses to point what they are talking about). They are clearly bored out of their minds the poor people!
Pinned article: Introduction to the OurBigBook Project
Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
Intro to OurBigBook
. Source. We have two killer features:
- topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculusArticles of different users are sorted by upvote within each article page. This feature is a bit like:
- a Wikipedia where each user can have their own version of each article
- a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
Figure 1. Screenshot of the "Derivative" topic page. View it live at: ourbigbook.com/go/topic/derivativeVideo 2. OurBigBook Web topics demo. Source. - local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
- to OurBigBook.com to get awesome multi-user features like topics and likes
- as HTML files to a static website, which you can host yourself for free on many external providers like GitHub Pages, and remain in full control
Figure 3. Visual Studio Code extension installation.
Figure 4. Visual Studio Code extension tree navigation.
Figure 5. Web editor. You can also edit articles on the Web editor without installing anything locally.Video 3. Edit locally and publish demo. Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.Video 4. OurBigBook Visual Studio Code extension editing and navigation demo. Source. 
- Infinitely deep tables of contents:
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact