H2O AutoML error Test/Validation dataset has a non-categorical column which is categorical in the training data" on predict

Question

I have trained and saved my H2O AutoML model. after reloading, while I am using predict method, I am getting below error: java.lang.IllegalArgumentException: Test/Validation dataset has a non-categorical column 'response' which is categorical in the training data

I have not specified any encoding while model creation but I am getting this error now. Can anyone help me on this issue.

Any help will be highly appreciated.

Can you specify which version of H2O you are using? This looks like an old bug that was fixed. — Erin LeDell, Aug 16 '19 at 15:27
I am seeing a similar error on 3.26.0.3 java.lang.IllegalArgumentException: Test/Validation dataset has a categorical response column 'C30' with no levels in common with the model. If you remove the y label from the "test" set and pass to model.predict() then it works. Was this behavior changed recently ? I am attempting to go through this diff : https://github.com/h2oai/h2o-3/compare/4854053b2e1773e6df02e04895709f692ebf7088...9d4c43ef5bd420a49af6df5bda3e1f89590d6c52 — dparkar, Aug 27 '19 at 21:18
Hi Atul Were you able to get an answer on this? I am facing a similar issue. — Vikram Garg, Jul 15 '20 at 13:33

score 4 · Answer 1 · answered Mar 24 '21 at 08:35

4

This issue related is with new examples data in particular column that doesn't exist in traing set. I use parsing column types to numeric (or string) in this cases.

def _convert_h2oframe_to_numeric(h2o_frame, training_columns):
    for column in training_columns:
        h2o_frame[column] = h2o_frame[column].asnumeric()
    return h2o_frame

Remember to use this function for training and prediction process.

answered Mar 24 '21 at 08:35

CezarySzulc

1,849
1
14
30

A similar solution worked for me! – Renata Ghisloti Apr 21 '22 at 15:03

score 0 · Answer 2 · answered Feb 14 '23 at 22:21

Maybe a Little late, but this problem still ocurrs, specially if you have lots of columns, what I dit to solve this problem was:

H2O gives one of two possible messages:

Test/Validation dataset has a non-categorical column '<YOUR-COLUMN>' which is categorical in the training data

or

Test/Validation dataset has categorical column '<YOUR-COLUMN>' which is real-valued in the training data

So, what I did was to extract the column name from the message and convert the column according to the message in categorical or numeric.

so, my python code looks like this:

hf = h2o.H2OFrame(df)
transform = True
while transform:
    try:
        prediction = rf_model.predict(hf)
        transform = False
    except Exception as inst:
        err_msg = str(inst)
        tarr = err_msg.split('categorical')
        column = tarr[1].split("'")[1]
        if tarr[0][-1] == '-': # convert to categorical
            hf[column] = hf[column].asfactor()
            print(f'{column} converted to categorical')
        else: # convert to numeric
            hf[column] = hf[column].asnumeric()
            print(f'{column} converted to real-valued')

Hope it helps!

H2O AutoML error Test/Validation dataset has a non-categorical column which is categorical in the training data" on predict

2 Answers2