Skip to content

Latest commit

 

History

History
67 lines (47 loc) · 2.81 KB

README.md

File metadata and controls

67 lines (47 loc) · 2.81 KB

Image Classification Workshop

Objectives:

  1. Learn Machine Learning basics

  2. Learn TensorFlow basics

  3. Learn how to write an image classification model with Tensorflow

  4. Learn 5 typical methods to improve model

Instructions to deploy using Google Colab (recommended):

  1. Each individual model will a Open in Colab button at the top that takes you to Google Colab and automatically loads the Jupyter notebook

  2. Log-in to your Google account and make sure that you are connected to Google Compute Engine backend

  3. You can use and activate the free GPU:

    Edit > Notebook Settings > Hardware Accelerator: GPU

    Please Note: A GPU is not required for this project, this is completely optional.

  4. Run each cell sequentially 1-by-1 OR Run them all at once: Runtime > Run all

  5. Click here to START

Instructions to deploy in your local machine (requires docker installed):

  1. Clone the repo and go to the ImageClassificationWorkshop directory

    git clone https://github.com/JPedro2/ImageClassificationWorkshop.git
cd ImageClassificationWorkshop/

  2. Build Docker Image

    docker build -t create_workshop .

  3. Run Jupyter Notebook

    docker run -it --rm -p 8888:8888 -v $(log directory on host):/tf/notebooks/logs create_workshop

    Grab the notebook URL from the output:

    [I 19:22:44.981 NotebookApp] Writing notebook server cookie secret to /root/.local/share/jupyter/runtime/notebook_cookie_secret
jupyter_http_over_ws extension initialized. Listening on /http_over_websocket
[I 19:22:45.196 NotebookApp] Serving notebooks from local directory: /tf
[I 19:22:45.196 NotebookApp] The Jupyter Notebook is running at:
[I 19:22:45.196 NotebookApp] http://dd379526f0b8:8888/?token=b38128d704f1958b585252a9132feeea81c7a95dcc48cb34
[I 19:22:45.196 NotebookApp]  or http://127.0.0.1:8888/?token=b38128d704f1958b585252a9132feeea81c7a95dcc48cb34
[I 19:22:45.196 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 19:22:45.201 NotebookApp] 
   
   To access the notebook, open this file in a browser:
      file:///root/.local/share/jupyter/runtime/nbserver-1-open.html
   Or copy and paste one of these URLs:
      http://dd379526f0b8:8888/?token=b38128d704f1958b585252a9132feeea81c7a95dcc48cb34
   or http://127.0.0.1:8888/?token=b38128d704f1958b585252a9132feeea81c7a95dcc48cb34

  4. Open the 127.0.0.1 URL with the token to get started

  5. Click notebooks and start with model_1.ipynb and work your way through to model_4.ipynb

  6. Run each cell sequentially 1-by-1 OR Run them all at once: Cell > Run all