Some information and resources

  • IPython (interactive) and Jupyter(notebok for noting and sharing)
    • Good thing about Jupyter: integrate code and document(markdown/html/latex).

    Ex. $softmax(x_i) = \frac{\exp^{x_i}}{\sum_{k=1}^K \exp^{x_k}}$ - Jupyter shortcut keys:

  • Tensorflow and Virtual environment:

  • Using Jupyter under virtualenv:
    • Activate your virtual environment (on Mac)

      source <virtual_env_path>/bin/activate

      Note that you should see (your_virtualenv_name) before the prompt when the virtual environment is successfully activated. Here we use (virtualenv) as an example.

    • Install ipython kernel to the virtualenv

      (virtualenv) pip install ipykernel  # pip3 install ipykernel for python3
    • Create your kernel

      (virtualenv) python -m ipykernel install --user --name=<your_kernel_name>
    • Lauch Jupyter and then you can see the newly added kernel

      (virtualenv) jupyter notebook
  • Tensorflow tutorial
    • Official tutorial:
    • Vedios about tensorflow (Chinese):
  • Tensorflow validation (hello world)
import tensorflow as tf
hello = tf.constant('Hello, Tensorflow')
sess = tf.Session()
Hello, Tensorflow
  • slim:
  • pre-trained model
  • GAN