question

TunahanDENZ-6224 avatar image
0 Votes"
TunahanDENZ-6224 asked LanreBolomope-7959 commented

I got error in inference pipeline in Azure machine Learning notebook

101608-ekran-goruntusu-3.png


ı am learning machine learning on microsoft learn and ı have problem with inference pipeline
inference-pipeline


azure-machine-learningazure-machine-learning-inference
ekran-goruntusu-3.png (106.5 KiB)
· 1
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

I hope this helps someone else as this is coming late for you.

I had a similar problem that turned out to be caused by the Clean Missing data transformation. I had included the column ( in your case "price") in the data cleaning.

This I believe was also included in the transformation created in the inference pipeline, and this is an input to the "Apply Transformation" step. This is what created the error in columns expected by the Apply Transformation step

To resolve this,

  1. Remove the "price" column from the clean missing data step

  2. Submit the training run again

  3. Register the cleaning transformation dataset (you can save it as a new version of the older)
    . List item

  4. Update the inference pipeline

  5. Go to the Inference pipeline and delete the cleaning transformation. ( It still retains the older transformation)

  6. Add the new transformation with the dataset
    163116-image.png

  7. Submit inference pipeline for run



0 Votes 0 ·
image.png (84.2 KiB)
image.png (122.2 KiB)
ramr-msft avatar image
1 Vote"
ramr-msft answered ramr-msft edited

@TunahanDENZ-6224 Thanks for the question. Can you please add more details steps that you performed. If you have changed the schema of the incoming data to exclude the price field, you must remove any explicit use of this field in other modules.

Please follow the document to create a real-time inference pipeline.
https://docs.microsoft.com/en-us/azure/machine-learning/tutorial-designer-automobile-price-deploy#create-a-real-time-inference-pipeline-1

We are able to create inference pipeline without any errors as shown below.
101618-image.png



image.png (72.2 KiB)
· 3
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

hi,ı am following this site Microsoft learn inference pipeline


and ı am following this image
101844-ekran-goruntusu-4-li.jpg


but as you can see ı dont understand blue area?

0 Votes 0 ·

ı am using this columns

[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]

in "select columns in dataset" module so there is not "price" column in "select columns in dataset" module

according to error code: price column does not exist in dataset but price column exist in transformation
101826-ekran-goruntusu-7-li.jpg



ı think we should not use price column in anywhere but there is probably price column in transformation side ı think this is real problem but ı can not be sure

ı should understand blue area

0 Votes 0 ·

@TunahanDENZ-6224 Thanks for the details, Extra Apply transformation is not required. We have forwarded to the team to update the details on the learning module that you are referring.
In this sample, Web Service Input and Automobile price data (Raw) connect to the same downstream module, hence Web Service Input expect the same data schema as Automobile price data (Raw) and target variable column price is included in the schema. However, usually When you score the data, you won't know the target variable values. For such case, you can remove the target variable column in the inference pipeline using Select Columns in Dataset module. Make sure that the output of Select Columns in Dataset removing target variable column is connected to the same port as the output of the Web Service Intput module.

0 Votes 0 ·
MikeRichardson-3493 avatar image
0 Votes"
MikeRichardson-3493 answered

you can try keeping "price" column in the dataset in the first Box,
Enter Data manually: price,column2,col3,....col10
Select Columns in Dataset: price,column2,col3,....col9 (choose which columns to import, omit col10 for example) Though the learning path says to remove price column, try keeping it in.
And you may be interested in Edit Metadata asset box. It allows your model to import the columns and associate the features with the predicted price. So instead of just seeing a predicted price, you can output, what you fed the model. Overall, did you get it to work by starting over? That learning Path worked for me when I followed all steps. maybe these extra ideas will help.116571-azmlpricecolumn1.png


116563-azmlpricecolumn2.png


116572-azmlpricecolumn3.png


116483-azmlpricecolumn4.png



5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.