TypeError: an integer is required about pymrmr

[shm-smn]

Did you find the solution for that? as i am encountering the same problem.

[fbrundu]

Hi @Kein0112,
Could you provide me a test matrix that generates such error so I can debug it?
Thanks,
Francesco

[fbrundu]

@shm-smn could you share a test matrix to understand the issue?

Thanks

[Irving-ren]

@fbrundu I just occurred the same problem. Do you familiar with this problem?

[fbrundu]

@terminator9487 could you share some test data to understand the issue?

Thanks

[Irving-ren]

Of course, Could I have you email address?

[fbrundu]

If it's not private you can just drop it here. Can you anonymize it?

[AlvaroRodriguezDelRio]

Hi! I am experiencing the same problem when inputing this matrix (which is a subset of a larger one for which I get the same error):

disease g1 g2 g3
CTR 0.000482 0.0 0.0
CRC 0.003343 0.0 0.0
CRC 0.006969 0.0 0.0
CTR 0.000000 0.0 0.0

Any idea of what the problem could be? Thanks in advanced!

[noobyogi0010]

Hi @AlvaroRodriguezDelRio !
I looked at the error and got to know that it is caused by the file pymrmr.pyx
Upon more debugging I found that it's because in the source file pymrmr.pyx where the mRMR function is defined, there's a line as follows:

1. def mRMR(data, method, nfeats):
2. cdef vector[vector[int]] _data = data.copy().values

In the line 2, you can see that the values of the input data, that's, the actual CSV file is copied into a int cell of the 2d vector. On the other hand, the data file you input contains values in float format, due to which it is giving the TypeError.
I'll create a pull request after making the appropriate changes to the pymrmr.pyx file.
I hope this might help you.
Thanks

[AlvaroRodriguezDelRio]

Thanks a lot for your answer @noobyogi0010 ! This was really helpful, I will be looking forward to the new updated version with the changes you propose!