Jupyter notebooks in Python to assist the tutorial on basics of Deep Learning. The present notebooks are intended for educational reasons only. They are not be perfect, but keep been gradually improved. Apologies for possible mistakes and missing references.
It is important to mention that the Python codes were inspired by the "Tutorial TensorFlow" provided by TensorFlow Core by TensorFlow authors. Modifications have been made on the original code for educational reasons and Python compatibility. Further information and more tutorials are available at:
https://www.tensorflow.org/tutorials
and
https://github.com/tensorflow/examples
Throughout this tutorial, we will run some python scripts using Jupyter Notebook. You can either download the files from this Gitlab repository and copy them in a folder of your choice or, if you prefer, you can clone it by typing:
git clone https://gitlab.esrf.fr/jdasilva/tutorialdeeplearning.git
Once the files are in your computer, you can open Jupyter Notebook by:
- Windows: click on the Jupyter Notebook icon installed in the start menu.
- Linux/Mac OS: open the terminal and type jupyter notebook at the prompt.
Concerning the use of Jupyter Notebooks, the website:
https://jupyter-notebook-beginner-guide.readthedocs.io/en/latest/index.html
shows how to install it and to start it in the different Operational Systems (Linux/Mac OS, Windows). Additionally, some Python packages are required: Numpy, IPython, Matplotlib, and scikit-image.
- python >= 3.6
- numpy
- matplotlib
- scipy
- Tensorflow
- Tensorflow-datasets
- Tensorflow_hub
- Keras
If you do not have such packages installed, I recommend the installation via pip install:
pip3 install --user numpy, ipython, matplotlib, scipy, tensorflow, tensorflow-datasets, tensorflow_hub, keras