This category is strictly for discussing topics that we have brought up in class, questions about running the code that we’ve looked at in class, or any other issues related to the lessons we’ve covered already.
The goal is for this category to be approachable and non-intimidating to folks with the minimum prerequisites (no machine learning background, one year of coding experience). In particular, posts should not cover future lessons, topics that aren’t in this course, setting up or using platforms other than AWS, Crestle, or Paperspace, or other topics outside of what we’ve covered in the lessons.
If you have a suggestion for other blog posts, books, etc which specifically cover a topic we’ve looked at in class in a way that is friendly to beginners, feel free to mention it when appropriate!
I have created a thread for each lesson: if you have any questions or comments about that lesson, please use that thread. Feel free to create new threads if your question or comment is more general, or covers some foundational issue like logarithms, ssh, numpy, etc.
Fantastic and thank you! I find the advanced topics interesting & inspirational. However, it is time-consuming to wade through & filter before discovering the threads that are best suited for me at the time I need them.
I am really enjoying the course & appreciate you being so responsive to the needs of all learners regardless of their start place on the timeline. This educational model is exceeding my expectations in every category. It definitely makes me think about more useful applications for AI that will add layers (not replace) to human decision making.
Am also in full admiration for you in maintaining this level of focus and attention & patience despite some obviously significant distractions in your life. Warm wishes and again a THANK YOU.
This is very helpful even for folks who’ve had prior exposure to ML. There are times when I only want to cement my understanding of basic concepts/clarity class topics, and other times when I like to brainstorm/browse what other folks are doing.
@jeremy Currently I am trying to understand the theory vs implementation ie: I am just going through the papers which are already implemented in pytorch to understand the transformation of theory into a code. I started with the Spatial transformers in pytorch http://pytorch.org/tutorials/intermediate/spatial_transformer_tutorial.html. In which they mentioned about the affine matrix which is kind of hard to understand. Could you please explain this in your terms just like max pooling and relu?
Also, I like to know am I in a right direction towards the deep learning?
Thank you for this section Jeremy. I’ve found that the machine learning community can be very hostile, and having a welcoming place like this is great.
Hi Jeremy and all others involved with the courses. As a beginning developer this course thought me a lot; not only about deep learning but at all related subjects as well. Thanks a lot for sharing it and looking forward to the next courses as things really move fast!
Hi
I have a beginner’s question regards path assignment - lesson 1
I downloaded the dogs cats folder and located it on the following path in my local CP:
C:\fastai\fastai\fastai\data\dogscats
From location: (fastai) c:\fastai\fastai>
On jupyter notebook, I run
PATH = “data/dogscats/”
BUT os.listdir(PATH) Shows only
[‘models’]
instead of dogscats folders
I have just started the course and completed dogscats example. Just for fun, i have tried with different datasets:
Cricket and baseketball.
Initially i have taken 10 photos each in training data and 5 in validation dataset. Rest details are same as class. My accuracy comes to ~46%. I than tried with more photos in training set (~30), accuracy improved to ~53. Still not good. Further i have increased set to 50 and move some photos to valid set as well. Now my accuracy reduced to 48%. I am confused what i am doing wrong…
Just one more thing, i changed the learning rate. For different learning rate my accuracy varied. The best i got is 70% with learning rate .003.
Any idea what should i try further to improve the accuracy?
I just watched the first lesson and was trying to get supporting materials and was very confused. I almost asked here, but found some of the answers myself and thought I’d share here, where hopefully the next lost soul can find it. If this is a bad spot and this post gets deleted or moved, please let me know.
The first place I went for help was to the wiki. It seems to be out of date and refer almost exclusively to the first version of the course.
Next I looked through the forums. Eventually I signed up and a blue dialog said, among other things, that there is a slack group, and that a hidden post on how to sign up for it would be revealed once I’d read at least three different posts on the forums for ten minutes all together. I like lurking in chat groups, so this seemed perfect. I poked around for a while and then started looking for it but couldn’t find it. After many frustrating minutes, I read the forum categories again. For some reason I’d unconsciously expected the hidden post to be at the top, in or near this forum category. It was the second from the last forum category, “Trust Level 1”, between the “Site Feedback” category and the “fastai dev” category. A stickied post in the previously hidden category said the Slack channel was no longer in use, so the forums and wiki are the place to go.
Edit: after more poking around, this looks like exactly what I wanted: Wiki: Lesson 1. The forum software itself has wiki capabilities.
Above is my data, when I am trying to use fastai, it report the error that “RuntimeError: Could not infer dtype of NoneType”, how can I fix it? Thank you.
