WARNING:nimare.utils:Argument <class 'nimare.meta.kernel.MKDAKernel'> has already been initialized, so arguments will be ignored: r, memory, memory_level

[JohannesWiesner]

Summary

I am getting this warning when initializing an instance of nimare.meta.kernel.MKDAKernel:

WARNING:nimare.utils:Argument <class 'nimare.meta.kernel.MKDAKernel'> has already been initialized, so arguments will be ignored: r, memory, memory_level

Additional details

• NiMARE version: 0.4.0

What were you trying to do?

Follow the tutorial up to this point (I also provided a memory instance):

# emacs: -*- mode: python-mode; py-indent-offset: 4; tab-width: 4; indent-tabs-mode: nil -*-
# ex: set sts=4 ts=4 sw=4 et:
"""

.. _metas5:

========================================================
 Run a meta-analytic coactivation modeling analysis
========================================================
Meta-analytic coactivation modeling (MACM) is a common coordinate-based
analysis in which task-independent "connectivity" is assessed by selecting
studies within a larger database based on locations of report coordinates.
"""
import os

import numpy as np
import nibabel as nib
from nilearn.image import load_img
from nilearn import plotting, image, datasets

import nimare
from nimare.tests.utils import get_test_data_path

###############################################################################
# Load Dataset
# --------------------------------------------------
# We will assume that the Neurosynth database has already been downloaded and
# converted to a NiMARE dataset.
dset_file = '/zi/home/johannes.wiesner/work/project_mase/data/neurosynth_dataset_with_abstracts.pkl.gz'
dset = nimare.dataset.Dataset.load(dset_file)

###############################################################################
# Define a region of interest
# --------------------------------------------------
# We'll use the right amygdala from the Harvard-Oxford atlas
atlas = datasets.fetch_atlas_harvard_oxford('sub-maxprob-thr50-2mm')

# FIXED THIS BECAUSE NIB GAVE AN ERROR
img = load_img(atlas['maps']) 

roi_idx = atlas['labels'].index('Right Amygdala')
img_vals = np.unique(img.get_fdata())
roi_val = img_vals[roi_idx]
roi_img = image.math_img('img1 == {}'.format(roi_val), img1=img)

###############################################################################
# Select studies with a reported coordinate in the ROI
# ----------------------------------------------------
roi_ids = dset.get_studies_by_mask(roi_img)
dset_sel = dset.slice(roi_ids)
print('{}/{} studies report at least one coordinate in the '
      'ROI'.format(len(roi_ids), len(dset.ids)))

###############################################################################
# Select studies with *no* reported coordinates in the ROI
# --------------------------------------------------------
no_roi_ids = list(set(dset.ids).difference(roi_ids))
dset_unsel = dset.slice(no_roi_ids)
print('{}/{} studies report zero coordinates in the '
      'ROI'.format(len(no_roi_ids), len(dset.ids)))


###############################################################################
# MKDA Chi2 with FWE correction
# --------------------------------------------------
mkda = nimare.meta.cbma.mkda.MKDAChi2(kernel__r=10)
mkda.fit(dset_sel, dset_unsel)

[JohannesWiesner]

Can solve this by creating an instance of kernel = MKDAKernel() and passing this to MKDAChi2(kernel_transformer=kernel). But the docs of MKDAChi2 say:

Keyword arguments. Arguments for the kernel_transformer can be assigned here, with the prefix 'kernel__' in the variable name.

so this should actually not be necessary?

[jdkent]

thank you for the report @JohannesWiesner!

I found the issue and have created a pull request in #907.

[jdkent]

this is fixed and released in 0.4.1, thanks again!