Thank you, Mike. This video is a great addition to the Azure AI Fundamentals course, especially for the module on the fundamentals of machine learning.

Thanks Mike

no preview datga buttom at the score and evaluate model boxes

Great way of explaining that concept

@mwcloud

5 ай бұрын

Thank you

Do you know how Can I upload my own Dataset? When I try to upload using the Data option in the asset section, it uploads a Data Asset not a dataset, any idea?

Thanks for the video and the explanation. Very useful. May I ask how to deploy the model in an interactive web app so different users can interact with it via slicers or buttons??

@mwcloud

2 ай бұрын

You would need to build your front end to interact with the model on the back end via an API call to the model. Depending on the load you may need the model deployed in a clustered scenario such as a k8s cluster and then load balance access to it

Thank you so much for sharing

@mwcloud

4 күн бұрын

You are so welcome

thank you .....

@mwcloud

Ай бұрын

You're most welcome

Do you wait for long time on select Columns in Dataset? I have waited for 15 minutes and it is still running on selecte columns in Dataset

Hi can you post the test data and Python script? Thanks

Any chance you could just paste the CSV you used at 34:40? You posted the URL where to get the data, and with limited experience with Python, it was a hassle to get the data. I did manage to extract it eventually, but even then, it seems not all columns work out of the bat. I selected a small subset of data, and it gives me this when it comes to scoring the model: azureml.studio.common.error.NotCompatibleColumnTypesError: Columns 'num-of-cylinders,num-of-doors' are incompatible between train and test data. I suppose this could be fixed with simple normalization of the data in between that step and the selection of columns, but I think it would be good to go over that. Otherwise it was a very good video and very informative!

@mwcloud

2 ай бұрын

I will do you one better, learn.microsoft.com/en-us/azure/machine-learning/tutorial-designer-automobile-price-train-score?view=azureml-api-1 have a look at the write up for this demo on the Machine Learning docs

Excuse me, in min 34:40 , when you input data from csv, where/how did u get that data? i guess is downloaded from before? but didnt understand where nor what nor how to download

@mwcloud

3 ай бұрын

archive.ics.uci.edu/dataset/10/automobile?darkschemeovr=1

@Minidrako99

3 ай бұрын

@@mwcloud thanks!

I've been trying to follow your tutorial, but after the first Data Transformation, such as Select Columns, it is impossible to make the columns list appera on a second transformation such as Cleaning Missing Data. I just keep getting the message: Autosuggestion only works if upstream node is a data node and schema exists for that data. The modules are connected, but it does not work. In classic Machine Learning Studio it worked smoothly. Also, if I used and runned Select Columns, I could preview the new dataset, wich I can no longer.

@mwcloud

4 ай бұрын

There could be a change that Microsoft has implemented since I have made this video, I am not sure without looking into detail about what you have built

How can I troubleshoot an unhealthy endpoint?

@mwcloud

9 ай бұрын

Please see this article, learn.microsoft.com/en-us/azure/machine-learning/how-to-troubleshoot-online-endpoints?view=azureml-api-2&tabs=cli I would recommend downloading the container logs as a good starting point to give you some clues

Hey Mike, can you please upload the that notepad where you have entered raw data again at 34:55 as per the timeline. I am following all your steps, just missing the data.

@animedemon6383

Ай бұрын

symboling, normalized-losses, make, fuel-type, aspiration, num-of-doors, body-style, drive- wheels, engine-location, wheel-base, length, width, height, curb-weight, engine-type, num-of- cylinders, engine-size, fuel-system, bore, stroke, compression-ratio, horsepower, peak-rpm, city- mpg,highway-mpg 3, NaN, alfa-romero, gas, std, two, convertible, rwd, front, 88.6,168.8,64.1,48.8,2548, dohc, four, 130, mpfi,3.47,2.68 ,9,111,5000,21,27 3, NaN, alfa-romero, gas, std, two, convertible, rwd, front, 88.6,168.8,64.1,48.8,2548, dohc, four, 130, mpfi,3.47,2.68 ,9,111,5000,21,27 1, NaN, alfa-romero, gas, std, two, hatchback, rwd, front,94.5,171.2,65.5,52.4,2823, ohcv, six,152,mpfi,2.68,3.47,9, 154,5000,19,26 might need to check once before using

@mwcloud

4 күн бұрын

Look through the comments, someone else already asked that

Hi we are testing the performance of API, it take avg 700ms for one call, we try to add more cpu/memory. Do you have any idea how to upscale?

@mwcloud

4 ай бұрын

Maybe it’s not the API, it could be your path to the API. Maybe try Azure front door with your solution. Hard to say really without looking at your deployment in detail

hi can i know how much does it cost to do the whole process

@mwcloud

4 ай бұрын

That whole thing? A few cents, nothing really as once it’s trained you are only really paying when it’s is processing data.

I unable to preview the data can anyone help

@chinwevivianaliyu

15 күн бұрын

same here

@abhin8444

9 күн бұрын

@@chinwevivianaliyu in current version of azure you can only preview tabular data and the microsoft azure has data in json format so you can not preview tabular data but you can try by adding custom data set in tabular format and then you can preview the data and I suggest you to take bike rentals or iris data set for better understanding and these datasetes will be available in the kaggle

YOU FORGOT TO ENCODE YOUR CATEGORICAL COLUMNS.

@mwcloud

3 ай бұрын

Thank you, but no need to shout

@praslisa

2 ай бұрын

@@mwcloud how do you encode? thanks this video! its fantastic!!

Great vídeo but such a psycho way of writing an "x" 2:17

@mwcloud

3 ай бұрын

😂