• OS: Ubuntu 20.04
• PyQt version: pyqt5==5.10.1

It would be nice to mention the dependency on libxml2-dev libxslt1-dev in the docs when building from source on Ubuntu.

I understand that these two python files show two different methods to construct a model. The original n_epoch is 500 which works perfect for both python files. But if I change n_epoch to 20, only tutorial_mnist_mlp_static.py can achieve a high test accuracy (~0.97). The other file tutorial_mnist_mlp_static_2.py only get 0.47.

The models built from these two files looks the same for me (the s

→ Step 1: Describe the issue 📝

bodypix returns error in p5js instance mode

→ Step 2: Screenshots or Relevant Documentation 🖼

→ Step 3: Share an example of the issue 🦄

https://editor.p5js.org/micuat/sketches/PYuow4Be3

• Web browser & version: chrome (but perhaps any)
• Operating System: windows (but perhaps any)
• ml5 version you're using: 0.5.0
• **Any addit

Thank you for the great effort, you are putting into this project :) There is, however, a feature I miss; rotated bounding boxes. Especially when objects are thin and diagonal, an ordinary bounding box fits poorly. Examples of such cases are shown here: rotated bounding boxes

A way annotation could be

Create text classification project that shows text to the annotator. The text also includes URLs or terms that the annotator may want to look up.
However normal copying the string to look up is not possible, presumably because the code that normally handles text selection in order create a sequence annotation (like for NER) is active.
"Normal copying" would be that the user marks the text to

We have a lot of repetitive instances in our data and it a simple Copy/Cut - Paste (Ctrl+C/X and Ctrl+V) function for bounding boxes would help speed up labelling by a lot!

Edit: Also multi-selection of multiple boxes/geometries would be great for a faster workflow

Modify 2_one-pixel-attack-imagenet.ipynb to download a sample subset of ImageNet files with appropriate labels. The files should be preprocessed as to be ready for the attack.

We need to create a separate dependencies list for react usage. Many react users won't need youtube-dl or ffmpeg libaries and we don't want things to be super bloated if they're using it as an npm module.

Describe the bug

Guide under Tensorflow.js heading needs an update, 'names' flag changed to 'name_policy ' according to latest commit?

Under supported layers heading, ConvTrsnpose2d==ConvTranspose2d?

Have you changed the basic TENSORBOX to run on multiclass? If you did, do you have some documentation on how to use your updated version of TENSORBOX with multiclass? If you didn't, on which network do you base your multiclass detection, is it YOLO? And what was the purpose of using TENSORBOX?
Thank you...

Python has a test package which can shadow the import of the test.py module, leading to an ImportError when importing NN or kNN.

https://docs.python.org/2.7/library/test.html?highlight=test#module-test
https://docs.python.org/3.5/library/test.html

I suggest to rename it to test_embedding.py or something like that.