Skip to content
Branch: master
Find file History
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
.ipynb_checkpoints Update 04-Deploy-Trained-Model-checkpoint.ipynb Aug 16, 2019
02-mobilenet-transfer-learning-scripts Rearrange folders (#188) Jun 19, 2019
aml_config Rearrange folders (#188) Jun 19, 2019
data/soda_cans Rearrange folders (#188) Jun 19, 2019
models Rearrange folders (#188) Jun 19, 2019
scripts Rearrange folders (#188) Jun 19, 2019
.library.json Rearrange folders (#188) Jun 19, 2019
00-aml-configuration.ipynb Merge pull request #236 from vidushv/MLnotebooks Aug 30, 2019
01-convert-model-containerize.ipynb Update 01-convert-model-containerize.ipynb Aug 30, 2019
02-mobilenet-transfer-learning.ipynb Update 02-mobilenet-transfer-learning.ipynb Aug 16, 2019
03-squeezenet-custom-vision.ipynb
04-Deploy-Trained-Model.ipynb Update 04-Deploy-Trained-Model.ipynb Aug 30, 2019
Dockerfile Rearrange folders (#188) Jun 19, 2019
README.md Update README.md Aug 30, 2019
camera.py Rearrange folders (#188) Jun 19, 2019
deployment-template.json Rearrange folders (#188) Jun 19, 2019
deployment.json Rearrange folders (#188) Jun 19, 2019
deploymentdemo.json Rearrange folders (#188) Jun 19, 2019
frame_iterators.py Rearrange folders (#188) Jun 19, 2019
iot.py Rearrange folders (#188) Jun 19, 2019
ipcprovider.py Rearrange folders (#188) Jun 19, 2019
main.py Rearrange folders (#188) Jun 19, 2019
utility.py Rearrange folders (#188) Jun 19, 2019
va-snpe-engine-library_config.json Rearrange folders (#188) Jun 19, 2019

README.md

Getting started with Jupyter Notebooks in Azure

Setup a NotebookVM in Azure Machine Learning Workspace.

  1. If you have an Azure Machine Learning service workspace, skip to step #2. Otherwise, create one now.

    • Sign in to the Azure portal by using the credentials for the Azure subscription you use.
    • In the upper-left corner of the portal, select Create a resource.
    • In the search bar, enter Machine Learning. Select the Machine Learning service workspace search result.
    • In the ML service workspace pane, scroll to the bottom and select Create to begin.
    • In the ML service workspace pane, configure your workspace and select Create. It can take a few minutes to create the workspace. When the process is finished, a deployment success message appears. It's also present in the notifications section. To view the new workspace, select Go to resource.
  2. Create a cloud-based notebook server.

    • Open your Machine Learning workspace in the Azure portal.
    • On your workspace page in the Azure portal, select Notebook VMs on the left.
    • Select +New to create a notebook VM.
    • Provide a name for your VM and select Create.
    • Wait approximately 4-5 minutes until the status changes to Running
  3. Launch the Jupyter wed interface in your Notebook VM

    • Select Jupyter in the URI column for your VM.
    • On the Jupyter notebook webpage, the top foldername is your username.

More details about quickstart setup instructions are located here.

Clone this repo to your Notebook VM

From the Notebook VM launch the Jupyter web interface as descriped in step #3 above. Click New -> Terminal on the upper right corner of the web interface. You will get a new browser tab with the bash prompt. You can use regular git clone --recursive https://github.com/microsoft/vision-ai-developer-kit command line commands to clone this repository into a desired folder.

Important update jupyter in the Notebook VM: (this is a temporary step)

pip install –upgrade notebook

sudo -i systemctl restart jupyter

Select a notebook in machine-learning-reference\notebooks to run it. Set the kernel to Python 3.6 - AzureML.

Notebooks in this repo

The 00-aml-configuration.ipynb contains the basic steps to setup the environment for Azure ML. This step is important to complete your config.json file

There are 4 notebooks in this folder as sample tutorials for the Vision AI Developer Kit. 01-convert-model-containerize.ipynb uses a pretained Mobilenet/Tensorflow model to convert and deploy on the Vision AI Dev Kit. 02-mobilenet-transfer-learning shows a transfer learning example to deploy the Mobilenet model on the Vision AI Dev Kit device. 03-squeezenet-custom-vision.ipynb converts a model that is imported from CustomVision.ai. 04-Deploy-Trained-Model.ipynb is an example to deploy a trained model to an existing IoT Edge device.

Also find other quickstarts and how-tos on the official documentation site for Azure Machine Learning service.

You can’t perform that action at this time.