Hi !
In my previous post I wrote about how to create a custom dataset with images to be used on a Azure Machine Learning Designer project. Today I’ll share the steps to use it.
Let’s start by creating a project based on the “Image Classification using DenseNet” template.

Next step is to select the compute instance we want to use in this Pipeline

Let’s delete the original animal datasource and add the newly created squirrel dataset.

Connect the new dataset with the [Convert to Image Directory] asset. To double check that we have the right contents, right-click on the Dataset asset and select [Preview Data], a list with the first 50 elements must appear on the preview window.

Now we can run the model. This may take a couple of minutes, so time for check twitter 😀

Once the pipeline run is done, we can check the score and metrics of the trained model. And, this is important for my online sessions, we can also download the trained model. Click on the [Train PyTorch Model] asset, select [Output and Logs] in the preview window, and we can download the file!

Tomorrow I’ll show how to create a HTTP Rest Endpoint to consume the generated model.
Happy coding!
Greetings
El Bruno
More posts in my blog ElBruno.com.
More info in https://beacons.ai/elbruno
¿Con ganas de ponerte al día?
En Lemoncode te ofrecemos formación online impartida por profesionales que se baten el cobre en consultoría:
- Si tienes ganas de ponerte al día con Front End (ES6, Typescript, React, Angular, Vuejs…) te recomendamos nuestros Máster Front End: https://lemoncode.net/master-frontend#inicio-banner
- Si te quieres poner al día en Backend (stacks .net y nodejs), te aconsejamos nuestro Bootcamp Backend: https://lemoncode.net/bootcamp-backend#bootcamp-backend/banner
- Y si tienes ganas de meterte con Docker, Kubernetes, CI/CD…, tenemos nuestro Bootcamp Devops: https://lemoncode.net/bootcamp-devops#bootcamp-devops/inicio
I created a new data set as described in part 1 of this blog. I cannot add this new dataset to the DenseNet sample project – the new data set does not appear in the list of available data sets.
The weird thing is I can add the new data set to a new custom project – the new dataset DOES appear if I start a new custom dataset. from scratch.
I have waited 3 hours hoping it would appear in the sample project but it does not. Any suggestions?
LikeLiked by 1 person
I created a new data set as described in part 1 of this blog. I cannot add this new dataset to the DenseNet sample project – the new data set does not appear in the list of available data sets.
The weird thing is I can add the new data set to a new custom project – the new dataset DOES appear if I start a new custom dataset. from scratch.
I have waited 3 hours hoping it would appear in the sample project but it does not. Any suggestions?
If I create a new project from scratch then paste the data set into the sample DenseNet project it throws the error below – I searched and no one has posted this error message.
“Submit Pipeline failed – Invalid Graph contains node not support support asset while using v2 data assets”
I CAN run the sample DenseNet project
IS the current DenseNet sample project out of data and wont run with a new data set?
In fact, I can not see ANY sample data sets if I create a new project from scratch..
Is this set of blogs out of date and no longer valid?
Any idea why I am getting this error and why I can’t see any sample data sets in a custom project now?
LikeLiked by 1 person
I think I am starting to see what is happening – there is a “classic” designer and a new custom designer for projects.
So here is the question – How do I make a data set of images as described in this blog that can be consumed by the “classic” designer. When I follow the instructions above, the data set is not available for use with the “classic” designer components – only the new custom designer
Is anything documented anywhere about the difference between classic datasets and the new v2 data sets and how to when to use each and create each data set type? If not, that would be a very good blog…
Thx
LikeLiked by 1 person
It’s been +1 year since this post. I may need to check the current statis of the Azure ML Designer.
I’ll try to update this in the near future !
LikeLike