czarrar / connectir

Fast and memory efficient approach for connectome-wide association studies in R and C++. Uses custom implementation of multivariate distance matrix regression to associate brain and behavior.
15 stars 9 forks source link

Error encountered when calling `connectir_subdist.R`: CDestroyBigMatrix not available #4

Closed neerajt closed 8 years ago

neerajt commented 8 years ago

When calling connectir_subdist.R on sample data, as follows;

connectir_subdist.R -i functional_path_list.txt --bg test_space.nii.gz --automask1 test_output

encountered the following error:

An error was detected: 
"CDestroyBigMatrix" not available for .Call() for package "bigmemory" 

Called by: 
.Call("CDestroyBigMatrix", x@address, PACKAGE = "bigmemory")

Saving options...

Removing everything from memory
...sucesss

Contents of functional_path_list.txt (points to test data included with package):

data/test_func01.nii.gz
data/test_func02.nii.gz
data/test_func03.nii.gz
data/test_func04.nii.gz
data/test_func05.nii.gz
data/test_func06.nii.gz
data/test_func07.nii.gz
data/test_func08.nii.gz
data/test_func09.nii.gz

Full output:

Loading niftir version < 1.0.

Setting 1 parallel forks
Setting 1 threads for matrix algebra operations
...using Intel's MKL
Checking required inputs
Checking optional inputs
...NOT using 2nd set of functional images
Preparing functional information

Preparing functionals
Preparing functional info
Preparing functional information

Computing brain mask
Calculating overlap of masks
...detecting file type
...reading in functionals and automasking
  |======================================================================| 100%
...checking for NaNs
...calculating overlap
...creating overlap mask
mask has 83.33% of non-zero voxels
final mask has 83.33% non-zero voxels
Memory?
Determining memory demands
...0.00 GB used for all functional data
...0.0 MB used for 2 distance matrices
...0.0 MB used for 2 connectivity maps across 9 subjects
...memory limit is 4.00 GB and a minimum of 0.00 GB is needed
...setting super-block size to 20 (out of 20)
...setting block size to 20 (out of 20)
...0.00 GB of RAM will be used
Creating output directory and files
...creating directories in /home/eta/MDMR/data/test_output
...creating soft links to subject functional data
...saving masks
s
s2
...copying background image
...saving options
...creating file-backed distance matrices
Loading and masking functional data (Part 1)
Reading in functional data
  |======================================================================| 100%
Scaling functional data
  |======================================================================| 100%
Checking input functionals (1)
Checking functionals
checking data
  |======================================================================| 100%
checking data
  |======================================================================| 100%
Setup is done. It took: 0.02 minutes

Computing subject distances
will run through 1 large blocks
large block 1
...creating temporary distance matrices
...compute distances
...running a test on first seed
...running a test on last seed
...now the real deal with 1 blocks and 18 seeds
  |======================================================================| 100%
...saving distances

An error was detected: 
"CDestroyBigMatrix" not available for .Call() for package "bigmemory" 

Called by: 
.Call("CDestroyBigMatrix", x@address, PACKAGE = "bigmemory")

Saving options...

Removing everything from memory
neerajt commented 8 years ago

Thank you. This is now resolved.