microsoft/ignite-learning-paths-training-aiml

Microsoft Ignite Learning Path, Train the Trainer materials: Developers Guide to AI

Please note: This repository is no longer actively maintained and therefore cannot be guarenteed that code and instructions are the latest information. To find out more about Azure AI services we recommend visting the documentation

Ignite Learning Paths - Developers Guide to AI

Welcome!

The content of this repository is available for you so you can reproduce any demo or learn how to present any session of the Learning Path presented at Microsoft Ignite and during Microsoft Ignite The Tour, in your local field office, a community user group, or even as a lunch-and-learn event for your company.

Do the Demos

If you are here to reproduce a demo in the comfort of your home/office, go in in the section Sessions. In each session you will find deployment instructions, to create the environment you need, and a tutorial to do the demo step by step.

Presenting the content

We're glad you are here and look forward to your delivery of this amazing content. As an experienced presenter, we know you know HOW to present so this guide will focus on WHAT you need to present. It will provide you a full run-through of the presentation created by the presentation design team.

Along with the video of the presentation, this repository will link to all the assets you need to successfully present including PowerPoint slides and demo instructions & code.

We are looking forward to working with all speakers who will deliver the content built below - we welcome your feedback and help to keep the content up-to-date.

Learning Path Description

Artificial Intelligence (AI) is driving innovative solutions across all industries but with machine learning (ML) applying a paradigm change to how we approach building products we are all exploring how to expand our skill-sets

Tailwind Traders is a retail company looking for support on how to benefit from applying AI across their business. In 'Developers Guide to AI’ we’ll show how Tailwind Traders has achieved this

There is something for every stage of the AI learning curve; whether you want to consume ML technologies, increase technical knowledge of ML theory, or build your own custom ML models. The model is not the end of the data science story, we will conclude with applying DevOps practices to ML projects to build an end-to-end pipeline

Sessions

• There are x5 45 minute sessions (AIML10, AIML20, AIML30, AIML40 and AIML50)
• And x1 20 min theatre session (AIML21)

Here all the sessions available in the learning path Developers Guide to AI (aka: AIML)

AIML10: Making Sense of your Unstructured Data with AI

Tailwind Traders has a lot of legacy data that they’d like their developers to leverage in their apps – from various sources, both structured and unstructured, and including images, forms, pdf files, and several others. In this session, you'll learn how the team used Cognitive Search to make sense of this data in a short amount of time and with amazing success. We'll discuss tons of AI concepts, like the ingest-enrich-explore pattern, skillsets, cognitive skills, natural language processing, computer vision, and beyond.

AIML20: Using Pre-Built AI to Solve Business Challenges

As a data-driven company, Tailwind Traders understands the importance of using Artificial Intelligence to improve business processes and delight customers. Before investing in an AI team, their existing developers were able to demonstrate some quick wins using pre-built AI technologies. In this session, we will show how you can use Azure Cognitive Services to extract insights from retail data and go into the neural networks behind computer vision. You’ll learn how it works and how to augment the pre-built AI with your own images for custom image recognition applications.

AIML30: Start Building Machine Learning Models Faster than You Think

Tailwind Traders uses custom machine learning models to fix their inventory issues – without changing their Software Development Life Cycle! How? Azure Machine Learning Visual Interface. In this session, you’ll learn the data science process that Tailwind Traders’ uses and get an introduction to Azure Machine Learning Visual Interface. You’ll see how to find, import, and prepare data, select a machine learning algorithm, train and test the model, and deploy a complete model to an API. Get the tips, best practices, and resources you and your development team need to continue your machine learning journey, build your first model, and more.

AIML40: Taking Models to the Next Level with Azure Machine Learning Best Practices

Tailwind Traders’ data science team uses natural language processing (NLP), and recently discovered how to fine-tune and build a baseline models with Automated ML.

In this session, you’ll learn what Automated ML is and why it’s so powerful, then dive into how to improve upon baseline models, using examples from the NLP best practices repository. We’ll highlight Azure Machine Learning key features and how you can apply them to your organization, including low priority compute instances, distributed training with auto scale, hyperparameter optimization, collaboration, logging, and deployment.

AIML50: Machine Learning Operations – Applying DevOps to Data Science 

While many companies have adopted DevOps practices to improve their software delivery, these same techniques are rarely applied to machine learning projects. Collaboration between developers and data scientists can be limited and deploying models to production in a consistent and trustworthy way is often a pipedream.

In this session, you’ll learn how to apply DevOps practices to your machine learning projects using Azure DevOps and Azure Machine Learning Service. We’ll set up automated training, scoring, and storage of versioned models and wrap the models in docker containers and deploy them to Azure Container Instances and Azure Kubernetes Service. We’ll even collect continuous feedback on model behavior so we know when to retrain.

AIML21: Developers Guide to AI: A Data Story

In this theatre session we will show the data science process and how to apply it. From exploration of datasets to deployment of services - all applied to an interesting data story. This will also take you on a very brief tour of the Azure AI Platform.

Contributing

To know more about contributing to this project please refer to the Code of Conduct and Contributing page.

Legal Notices

Microsoft and any contributors grant you a license to the Microsoft documentation and other content in this repository under the Creative Commons Attribution 4.0 International Public License, see the LICENSE file, and grant you a license to any code in the repository under the MIT License, see the LICENSE-CODE

Microsoft, Windows, Microsoft Azure and/or other Microsoft products and services referenced in the documentation may be either trademarks or registered trademarks of Microsoft in the United States and/or other countries. The licenses for this project do not grant you rights to use any Microsoft names, logos, or trademarks. Microsoft's general trademark guidelines can be found at http://go.microsoft.com/fwlink/?LinkID=254653.

Privacy information can be found at https://privacy.microsoft.com/en-us/

Microsoft and any contributors reserve all other rights, whether under their respective copyrights, patents, or trademarks, whether by implication, estoppel or otherwise.

ignite-learning-paths-training-aiml's Issues

Incompatible tensorboard versions cause error - AIML40

I get the following error when running this notebook.
Should I make some changes in my env configuration ?

Successfully built nlp-architect sklearn
ERROR: tensorflow-gpu 2.1.0 has requirement tensorboard<2.2.0,>=2.1.0, but you'll have tensorboard 1.13.1 which is incompatible.
ERROR: tensorflow-gpu 2.1.0 has requirement tensorflow-estimator<2.2.0,>=2.1.0rc0, but you'll have tensorflow-estimator 1.13.0 which is incompatible.
ERROR: azureml-train-automl-runtime 1.0.83.1 has requirement pandas<=0.23.4,>=0.21.0, but you'll have pandas 0.24.2 which is incompatible.
ERROR: azureml-train-automl-runtime 1.0.83.1 has requirement scipy<=1.1.0,>=1.0.0, but you'll have scipy 1.4.1 which is incompatible.
ERROR: azureml-train-automl-runtime 1.0.83.1 has requirement wheel==0.30.0, but you'll have wheel 0.33.6 which is incompatible.
ERROR: azureml-opendatasets 1.0.83 has requirement pandas<=0.23.4,>=0.21.0, but you'll have pandas 0.24.2 which is incompatible.
ERROR: azureml-opendatasets 1.0.83 has requirement scipy<=1.1.0,>=1.0.0, but you'll have scipy 1.4.1 which is incompatible.
ERROR: azureml-defaults 1.0.83 has requirement flask==1.0.3, but you'll have flask 1.1.1 which is incompatible.
ERROR: azureml-datadrift 1.0.83 has requirement matplotlib==3.0.2, but you'll have matplotlib 3.1.2 which is incompatible.
ERROR: azureml-automl-runtime 1.0.83 has requirement azureml-automl-core==1.0.83, but you'll have azureml-automl-core 1.0.83.1 which is incompatible.
ERROR: azureml-automl-runtime 1.0.83 has requirement pandas<=0.23.4,>=0.21.0, but you'll have pandas 0.24.2 which is incompatible.
ERROR: azureml-automl-runtime 1.0.83 has requirement scipy<=1.1.0,>=1.0.0, but you'll have scipy 1.4.1 which is incompatible.
ERROR: azureml-automl-runtime 1.0.83 has requirement wheel==0.30.0, but you'll have wheel 0.33.6 which is incompatible.
Installing collected packages: tensorboard, tensorflow-estimator, tensorflow, dynet, murmurhash, plac, cymem, preshed, wasabi, blis, srsly, thinc, spacy, sklearn, pandas, tensorflow-hub, nlp-architect

got unknown error on aiml50 demonstration setup

Hi

Need some help, I got below unknown error when try to deploy aiml50 demonstration. And I'm already follow through setup instruction... Please help...

"properties": {
    "statusCode": "BadRequest",
    "serviceRequestId": null,
    "statusMessage": "{\"error\":{\"code\":\"InvalidTemplateDeployment\",\"message\":\"The template deployment 'Microsoft.Template' is not valid according to the validation procedure. The tracking id is '2a482f22-cca8-4664-bb71-cc9e4ea74360'. See inner errors for details.\",\"details\":[{\"code\":\"ValidationForResourceFailed\",\"message\":\"Validation failed for a resource. Check 'Error.Details[0]' for more information.\",\"details\":[{\"code\":\"ServerFarmNotFound\",\"message\":\"The specified app service plan was not found.\"}]}]}}"
},

Thanks

The AIML40 data uploaded by the script is corrupt

After uploading the dataset for demo2 using the workaround detailed in #59, the data that is uploaded appears to be corrupt:

AIML20 vision_demo.sh link updates needed

In AIML20 vision_demo.sh file has links to old repo ignite-learning-paths rather than ignite-learning-paths-training-aiml since merge of repos. images linked are drill and man wearing hardhat

AIML40 bug 2

I found another problem in AIML 40 demo2, when I use clothing_automl.xlsx it to dataset , I found preview show error like this ，characters are garbled。 I use UTF-8 ..........

AML50 deployment storage account name already taken

Hello, could you please take a look at the Deploy to Azure template for AML50? I am getting deployment storage account name already taken error. I'm not sure how dependent the services are to this name so not sure if it's an easy single name fix in the template definitions.

westeurope location deployment is failing for TWT website ARM template (AIML20). But worked for westus2.

Just reproducing instructions from here. Using westeurope as a location.

Extra column in sample csv

There is an extra column at the end of https://github.com/microsoft/ignite-learning-paths-training-aiml/blob/master/aiml30/C%23/IgniteAimlDataApp/IgniteAimlDataApp/Datasets/ForecastingData.csv that causes some troubles

Please clarify Compute types required in demo

In blob/master/aiml30/README-Attendee.md please clarify the Compute type that should be created. The UI now separates Notebook VM / Training / Inference. I assume that the Kubernetes compute should be created under Inference for example...

Please clarify the path to the csv file in AIML30

I'm not sure what the 'correct path' will be for the uploaded csv file in the .ipynb.

In Datasets I see a relative path property for the file after uploading but I tried that and I get a FileNotFoundError.

AIML40: No compute available to profile dataset

In README.md where it says:

Click Next two times,0 on the final page select Profile the dataset after creating and click Create.

It is not possible to check "Profile the dataset" until a Notebook VM has been created, which has not yet happened at this point in the setup process.

AIML 40 BUG

When I run absa.ipynb

it show me this error :
Error occurred: User program failed with ModuleNotFoundError: No module named 'tensorflow.python.training.tracking'

Path

The AIML40 upload script gives an error

Running the command to upload the dataset:

python upload_dataset.py -s [subscription_id] -w absa_space -g absa -f clothing_automl.xlsx

This gives the following error:

Traceback (most recent call last):
  File "upload_dataset.py", line 4, in <module>
    from azureml.core import Workspace,Dataset
ModuleNotFoundError: No module named 'azureml'

AIML10 - Relative Import error when running Reader Skill locally

Received this error while testing Demo 3 of the AIML demos, during local run:

A ScriptHost error has occurred
Exception while executing function: Functions.AnalyzeInvoice. Microsoft.Azure.WebJobs.Script: Traceback (most recent call last):
  File "D:\Internal\Trainings\ignite-learning-paths-training-aiml-master\aiml10\src\InvoiceReaderSkill\AnalyzeInvoice\func.py", line 7, in <module>
    from . import helpers
ImportError: attempted relative import with no known parent package
.

As far as I can tell I followed the instructions perfectly, which from my experience usually means I need to "down-date" one of the tools back to the version from when the demo was published. Any ideas?

Broken Link for Invoice Training Dataset

Great work on the documentation and really cool demos!

I noticed that the link found in demo2 page for the Invoice Training Set is broken.

It would be really helpful if you could update the link with a valid one.

The AIML40 docs don't explain the data file upload

In AIML40, the docs to upload the data file don't provide detailed instructions to cover what settings are needed for things like the column headers, which columns are needed etc.

Can these instructions be made more explicit please, especially as the upload script isn't working (see #60).