Hi I wrote a blog post about some nice tools for Jupyter Notebook. I am doing the machine learning course so I don’t know where I should put this thread. I wrote this blog mainly to summarize the things that I found useful, there are definitely more but I didn’t have a good habit to make notes before. I found by writing blogs it forces me to revise the material and have a better memory of it. Will keep adding when I start doing the v2 course. I hope this blog makes it easier to set up the extension, as I have been struggling with how to get that collapsible heading before and somehow the toolbar is gone.
I always found useful things in the course, that I have been doing in a much slower way. I never know there is %prun until I saw it in lectures.
Another nice post! Have you got a twitter handle? Would love to mention it when I tweet a link to this.
BTW for shell commands it’s maybe better to use ! instead of %. BTW I don’t generally recommend pep8 for interactive coding - see the FAQ on this forum for a brief explanation. Entirely my particular opinion of course, but just thought I’d mention it in case it’s of interest.
Thank you for writing a blog post to share how to install Jupyter extension package and some useful extension. There’s more useful Jupyter extensions in the following article. We can also learn some of the tricks and tips there as well.
Thanks. I mainly summarize what I have learned in this course, maybe I can write another post if I have experienced more useful tricks. The best thing I like in this course is other than the hard knowledge, I also learned a lot of little neat skills to enhance my workflow.
nbdime works well locally but can’t be used for GitHub PRs. They built a solution that uses GitHub APIs and makes it easy to review PR and write comments.
If you want to exclude cell outputs from the comparison and you don’t want to commit the outputs to your repository/version control, check out nbstripout.
Disclosure: I helped them during their product validation. No financial ties.
Sounds great, but I cannot use it for my work as we are not using Github, will keep an eye on it. I am actually surprise there are no online diff tool, Kaggle Kernel allows you compare different version of the same notebook(for input only) for quite a while already. So at least comparing notebook online is not impossible.