申し訳ありません、このコンテンツはただ今 アメリカ英語 のみです。 For the sake of viewer convenience, the content is shown below in the alternative language. You may click the link to switch the active language.

Environment

Q: What Python environment is available?

A: We are using the Anaconda2 distribution from Continuum Analytics, which is built around version 2.7 of the interpreter.

Q: What packages are already available?

A: Most popular packages for data science and machine learning are already installed and ready to be imported. The current list is available here.

Q: Why not use Python3?

A: At the moment most deep learning libraries written in Python focus on version 2.7.

How to use

Q: I modified a notebook and now it does not work well. How can I recover the notebook?

A: The procedure is as follows:

  1. go to the material page
  2. download the notebook file (*.ipynb) that you want to repair
  3. upload the file to the ipython notebook environment that you are using

Q: I wrote some codes on my ipython notebook environment last week and it’s gone now.

A: We replace all your environment every Sunday midnight. You should save your code to your local machine if it’s important.

Q: Where can I download the exercise materials ?

A: They can be download from https://github.com/ailorg/dl_lecture_pub

Q: What should I do if I run out of time?

A: If you need more time please use this form.

Homework

Q: Are there any rules for submission?

A: There are several things that you need to be careful about:

  • Complete the homework function.
    • “train_X” and “train_y”, and “test_X” are provided for training and testing respectively.
    • For validation, “train_X” and “train_y” can be divided into “valid_X” and “train_X”, and “train_y” and “valid_y”.
    • You need to return a variable “pred_y” according to the value of “test_X”
    • The lines starting from “%” will be removed.
    • The lines containing “test_y” will be removed.
  • Your score will be evaluated by F-measure.

Review the homework submission guide here.

Q: Can I change my submission?

A: Yes, it is possible to submit your homework as many times as you want during the submission period. Once it ends the last homework submitted will be evaluated.

Q: Are there any rules for evaluation?

A: Yes, the code must end without raising any errors and finish before the designated evaluation time (60 minutes).

Q: Why am I not in the leaderboard?

A: Only the ten best results get into the leaderboard. If your score was not high enough you will not be listed as a top entry.

Troubleshooting

In some cases, these tips might work.