To stay on top of the latest trends in machine learning, you need to be fast. Things change quickly and round the clock. And things happen even faster in computer vision. People and machines generate millions of visuals–videos, images, gifs, memes…
Take this– according to these statistics, Instagram users upload over 100 million photos and videos every day. That is 69,444 million posts every minute! So imagine how many pictures the algorithms need to process to gather all the data necessary for interpreting and understanding the visual world.
But let’s get to the point. With all that data, computer vision in the machine learning industry is rapidly progressing. So to help you stay on top of the latest trends, we’ve gathered the best resources. Here we go!
PyImageSearch is a website created by Adrian Rosebrock, PhD, professional Computer Vision/Deep Learning developer and researcher. He started the community to help other students, developers, and researchers become better at computer vision.
He shows other people how to master Computer Vision and Deep Learning in a practical way instead of diving into theoretical stuff
How to use it? Just take a look around the website and choose what you want to work on. You can choose from 4 types of resources:
- A step-by-step guide to getting started, getting good, and mastering Computer Vision, Deep Learning, and OpenCV.
- Go through different k topics depending on your interest
- Books and Courses written and created by Adrian
- Student Success Stories with several interesting case studies and reviews to check whether it’s what you need
- Blog with articles, guides, tutorials, or interviews
Old but gold, Reddit never disappoints! This thread is home to members interested or experienced in various fields from image processing, machine learning to signal processing. It helps people with their queries and shares useful and relevant information.
It’s real simple–use filters to quickly search for a specific topic or indulge yourself in infinite scrolling 😉 Just be careful, it’s easy to get lost in all the content.
Let’s stick with Reddit for a while. Here, you have a search for everything related to Computer Vision in the Machine Learning thread.
Use it to quickly find a specific question, query, or discussion. You can limit your search to Best results, Posts, Communities, and users. Quick but effective.
Google is one of the pioneers in using AI. So why not follow it to check the latest trends and information? It has all the latest news from Google AI and is regularly updated with new posts written by researchers at Google who share their work with the world.
Whether you need inspiration, are looking for specific information, or just want to check what’s happening at Google, you can freely search the blog for what you want. Use labels and Archive to narrow your hunt.
Here, you’ll find the insights from the Facebook Researchers who work to enhance people’s experiences across Facebook products. On the website, you can find the latest publications, articles, news, and videos about Computer Vision.
You can research people and reach out to them (the list is long!):
- go through publications,
- read the blog,
- sign up for events,
- or even find a job.
You can also use additional resources – downloads and projects, and check for visiting researchers and postdocs programs.
Tombone’s Computer Vision Blog has everything about Deep Learning, Computer Vision, and the algorithms that are shaping the future of Artificial Intelligence.
It’s an excellent resource for demanding. The author, a Senior Research Scientist at Amazon Robotics AI, publishes great content and shares his knowledge on the blog.
Although new content is published occasionally, it’s worth checking the blog on a regular basis. Either to read some of the old posts or wait for the new ones. It’s a simple blog with tons of helpful and interesting information.
An interesting blog written by an interesting person! It should be on your list of the best resources on Computer Vision and Machine Learning. The author (who has a PhD in Computer Vision and a Master’s in Theology and a Master’s in Philosophy!) writes about interesting things going on in the world of Computer Vision.
To give you a taste, here are the top 5 posts from the blog (according to the readers, and the author).
Just take a look at the blog and see if there’s something of interest for you.
Analytics Vidhya is a website where you can find all types of information on ML, including Computer Vision. In this category, you can read different in-depth, well-researched articles on the topic.
Make sure to check the website out for other interesting information, courses, and more!
Towards Data Science is a popular Medium community that helps thousands of people to exchange ideas and to expand the understanding of data science.
When it comes to Computer Vision, TDS has an entire category dedicated to this subject. Different authors contribute and share their knowledge and insights in gripping articles. Make sure to check it out as well as other categories. New articles appear every day!
Hackernoon is another helpful resource for enthusiasts of Computer Vision. Like other websites, it has a category dedicated to this subject. All articles are written by experts on the subject.
And don’t forget to go through the website to check other tech-related stories! It’s a great site for all ML lovers.
Who doesn’t know the famous KDNuggets! It’s one of the most popular sites where you can find all types of resources. From articles, news, to research papers, and jobs. The content is always fresh, engaging, and professional.
Tagged with ‘Computer Vision’ the site shows you all the relevant articles.
You can also sign up for their newsletter to stay up-to-date with the latest trends.
In our blog archives, you can find articles on Computer Vision written by highly experienced and skilled professionals. Check them out to learn new skills, deepen your knowledge, and discover how other people approach experimenting.
Always fresh and ready for you to pick it up as your next read! 😉
But that’s not all! You can find more information on our blog in different categories, go through resources, or even sign up for our tool to get your ML experiments under control!
To wrap it up
So as you can see, following the current trends is demanding.
There are lots of resources, lots of reading, and lots of practice. Bookmark your favorite sites and always stay on top!
ML Experiment Tracking: What It Is, Why It Matters, and How to Implement It
Jakub Czakon | Posted November 26, 2020
Let me share a story that I’ve heard too many times.
”… We were developing an ML model with my team, we ran a lot of experiments and got promising results…
…unfortunately, we couldn’t tell exactly what performed best because we forgot to save some model parameters and dataset versions…
…after a few weeks, we weren’t even sure what we have actually tried and we needed to re-run pretty much everything”
– unfortunate ML researcher.
And the truth is, when you develop ML models you will run a lot of experiments.
Those experiments may:
- use different models and model hyperparameters
- use different training or evaluation data,
- run different code (including this small change that you wanted to test quickly)
- run the same code in a different environment (not knowing which PyTorch or Tensorflow version was installed)
And as a result, they can produce completely different evaluation metrics.
Keeping track of all that information can very quickly become really hard. Especially if you want to organize and compare those experiments and feel confident that you know which setup produced the best result.
This is where ML experiment tracking comes in.Continue reading ->