Skip to content

Error when running the analysis on an image with odd width or height #4

New issue
New issue
Open
Open
Error when running the analysis on an image with odd width or height#4
@bpavie

Description

@bpavie
bpavie
opened on Apr 5, 2023

Hi,

I cropped the sample image and had an odd height, and tried to run it but get the following error:

$ python3 hairyroots.py -i sample/TAKFA3-n3-1_Classes_smaller.tif -o sample/ --id_root 3 --id_roothair 1 --id_background 2 --pixel_size 1.72 -p 
usage: hairyroots.py [-h] [-i INPUT_PATH] [-o OUTPUT_PATH] [-p] [--pixel_size PIXEL_SIZE] [--id_root ID_ROOT] [--id_background ID_BACKGROUND] [--id_roothair ID_ROOTHAIR] [--max_dist MAX_DIST] [--thresh_d2r THRESH_DIST_TO_ROOT]
                     [--prune IS_PRUNE] [--bin_op IS_CLOSE_GAPS] [--measure {strain_energy,total_curvature}] [--cost_type {mean,exp,rms,pow3,pow4,geom}] [--n_levels N_LEVELS] [--n_repeats N_REPEATS] [--use_ref_tips] [--no_ref_tips]
                     [--w_curve W_CURVE] [--w_len W_LEN] [--w_mind W_MIND]

Extracts and measures root hairs from classified image.

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT_PATH, --in INPUT_PATH
                        tiff input file
  -o OUTPUT_PATH, --out OUTPUT_PATH
                        csv output filename
  -p, --print           Select to output other data
  --pixel_size PIXEL_SIZE
                        pixel size in microns per pixel
  --id_root ID_ROOT     id of root
  --id_background ID_BACKGROUND
                        id of background
  --id_roothair ID_ROOTHAIR
                        id of root hairs
  --max_dist MAX_DIST   Maximum distance of root hair clusters to root in pixels.
  --thresh_d2r THRESH_DIST_TO_ROOT
                        Minimum distance of root hair to root
  --prune IS_PRUNE      Preprocessing step: Prune medial axis
  --bin_op IS_CLOSE_GAPS
                        Preprocessing step: Binary opening/closing
  --measure {strain_energy,total_curvature}
                        Type of curvature measure
  --cost_type {mean,exp,rms,pow3,pow4,geom}
                        Way to summarize optimization objectives.
  --n_levels N_LEVELS   Minimum number of iteration levels for optimization.
  --n_repeats N_REPEATS
                        Number of times simulated annealing is repeated.
  --use_ref_tips        Treat tips separately from non-tips for referance values.
  --no_ref_tips         Do not treat tips separately from non-tips for referance values.
  --w_curve W_CURVE     Weight for curvature optimzation.
  --w_len W_LEN         Weight for length optimzation.
  --w_mind W_MIND       Weight for minimum distance to root optimzation.
usage: hairyroots.py [-h] [-i INPUT_PATH] [-o OUTPUT_PATH] [-p] [--pixel_size PIXEL_SIZE] [--id_root ID_ROOT] [--id_background ID_BACKGROUND] [--id_roothair ID_ROOTHAIR] [--max_dist MAX_DIST] [--thresh_d2r THRESH_DIST_TO_ROOT]
                     [--prune IS_PRUNE] [--bin_op IS_CLOSE_GAPS] [--measure {strain_energy,total_curvature}] [--cost_type {mean,exp,rms,pow3,pow4,geom}] [--n_levels N_LEVELS] [--n_repeats N_REPEATS] [--use_ref_tips] [--no_ref_tips]
                     [--w_curve W_CURVE] [--w_len W_LEN] [--w_mind W_MIND]
Namespace(input_path='sample/TAKFA3-n3-1_Classes_smaller.tif', output_path='sample/', print_all=True, pixel_size=1.72, id_root=3, id_background=2, id_roothair=1, max_dist=10, thresh_dist_to_root=10, is_prune=True, is_close_gaps=True, measure='total_curvature', cost_type='rms', n_levels=1000, n_repeats=1, use_ref_tips=True, w_curve=1.0, w_len=1.0, w_mind=1.0, func=<function run_pipeline at 0x14614fab9160>)
memory use: 0.1386
Elapsed time: 00:00:00
memory use: 0.1578
Elapsed time: 00:00:00
Traceback (most recent call last):
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/hairyroots.py", line 579, in <module>
    main()
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/hairyroots.py", line 574, in main
    args.func(args)
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/hairyroots.py", line 84, in run_pipeline
    ma, ma_dist, dist_to_root, data = prep.run(data)
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/preprocessing.py", line 37, in run
    classes = self.removeFarRootHairs(classes, max_distance=self.max_distance, id_root=self.id_root, id_background=self.id_background, id_roothair=self.id_roothair)
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/preprocessing.py", line 122, in removeFarRootHairs
    edge = self.find_edge(img, id_root)       # pixel coordinates of edge of main root
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/preprocessing.py", line 292, in find_edge
    n_neighbours_edge = segmentation.Segmentation.numOfNeighbours(edge_loc, arr)
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/segmentation.py", line 86, in numOfNeighbours
    nneighbours = rank.sum(arrBinary, a)            # sum of neighbouring pixels
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/dirtmu_env/lib/python3.9/site-packages/skimage/_shared/utils.py", line 282, in fixed_func
    return func(*args, **kwargs)
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/dirtmu_env/lib/python3.9/site-packages/skimage/filters/rank/generic.py", line 1068, in sum
    return _apply_scalar_per_pixel_3D(generic_cy._sum_3D, image,
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/dirtmu_env/lib/python3.9/site-packages/skimage/filters/rank/generic.py", line 276, in _apply_scalar_per_pixel_3D
    image, footprint, out, mask, n_bins = _handle_input_3D(
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/dirtmu_env/lib/python3.9/site-packages/skimage/filters/rank/generic.py", line 198, in _handle_input_3D
    raise ValueError('Image dimensions and neighborhood dimensions'
ValueError: Image dimensions and neighborhood dimensions do not match

When I re-crop the image to none odd width/height, it works, but I get a different error at the end of the analysis:


$ python3 hairyroots.py -i sample/TAKFA3-n3-1_Classes_smaller.tif -o sample/ --id_root 3 --id_roothair 1 --id_background 2 --pixel_size 1.72 -p 
usage: hairyroots.py [-h] [-i INPUT_PATH] [-o OUTPUT_PATH] [-p] [--pixel_size PIXEL_SIZE] [--id_root ID_ROOT] [--id_background ID_BACKGROUND] [--id_roothair ID_ROOTHAIR] [--max_dist MAX_DIST] [--thresh_d2r THRESH_DIST_TO_ROOT]
                     [--prune IS_PRUNE] [--bin_op IS_CLOSE_GAPS] [--measure {strain_energy,total_curvature}] [--cost_type {mean,exp,rms,pow3,pow4,geom}] [--n_levels N_LEVELS] [--n_repeats N_REPEATS] [--use_ref_tips] [--no_ref_tips]
                     [--w_curve W_CURVE] [--w_len W_LEN] [--w_mind W_MIND]

Extracts and measures root hairs from classified image.

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT_PATH, --in INPUT_PATH
                        tiff input file
  -o OUTPUT_PATH, --out OUTPUT_PATH
                        csv output filename
  -p, --print           Select to output other data
  --pixel_size PIXEL_SIZE
                        pixel size in microns per pixel
  --id_root ID_ROOT     id of root
  --id_background ID_BACKGROUND
                        id of background
  --id_roothair ID_ROOTHAIR
                        id of root hairs
  --max_dist MAX_DIST   Maximum distance of root hair clusters to root in pixels.
  --thresh_d2r THRESH_DIST_TO_ROOT
                        Minimum distance of root hair to root
  --prune IS_PRUNE      Preprocessing step: Prune medial axis
  --bin_op IS_CLOSE_GAPS
                        Preprocessing step: Binary opening/closing
  --measure {strain_energy,total_curvature}
                        Type of curvature measure
  --cost_type {mean,exp,rms,pow3,pow4,geom}
                        Way to summarize optimization objectives.
  --n_levels N_LEVELS   Minimum number of iteration levels for optimization.
  --n_repeats N_REPEATS
                        Number of times simulated annealing is repeated.
  --use_ref_tips        Treat tips separately from non-tips for referance values.
  --no_ref_tips         Do not treat tips separately from non-tips for referance values.
  --w_curve W_CURVE     Weight for curvature optimzation.
  --w_len W_LEN         Weight for length optimzation.
  --w_mind W_MIND       Weight for minimum distance to root optimzation.
usage: hairyroots.py [-h] [-i INPUT_PATH] [-o OUTPUT_PATH] [-p] [--pixel_size PIXEL_SIZE] [--id_root ID_ROOT] [--id_background ID_BACKGROUND] [--id_roothair ID_ROOTHAIR] [--max_dist MAX_DIST] [--thresh_d2r THRESH_DIST_TO_ROOT]
                     [--prune IS_PRUNE] [--bin_op IS_CLOSE_GAPS] [--measure {strain_energy,total_curvature}] [--cost_type {mean,exp,rms,pow3,pow4,geom}] [--n_levels N_LEVELS] [--n_repeats N_REPEATS] [--use_ref_tips] [--no_ref_tips]
                     [--w_curve W_CURVE] [--w_len W_LEN] [--w_mind W_MIND]
Namespace(input_path='sample/TAKFA3-n3-1_Classes_smaller.tif', output_path='sample/', print_all=True, pixel_size=1.72, id_root=3, id_background=2, id_roothair=1, max_dist=10, thresh_dist_to_root=10, is_prune=True, is_close_gaps=True, measure='total_curvature', cost_type='rms', n_levels=1000, n_repeats=1, use_ref_tips=True, w_curve=1.0, w_len=1.0, w_mind=1.0, func=<function run_pipeline at 0x14ca644db160>)
memory use: 0.1429
Elapsed time: 00:00:00
memory use: 0.1534
Elapsed time: 00:00:00
cleaning up tips...
cleaning up tips...
cleaning up tips...
memory use: 0.2658
Elapsed time: 00:00:31
memory use: 0.3268
Elapsed time: 00:00:14
**************************************************
             Getting candidates                   
      # of connected components:80
**************************************************
Component 0:  3190 nodes
 - 313449 candidates; memory use: 0.5601
Component 1:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 2:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 3:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 4:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 5:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 6:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 7:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 8:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 9:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 10:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 11:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 12:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 13:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 14:  3 nodes
 - 1 candidates; memory use: 0.5604
Component 15:  1133 nodes
 - 96109 candidates; memory use: 0.6331
Component 16:  3 nodes
 - 1 candidates; memory use: 0.6331
Component 17:  3 nodes
 - 1 candidates; memory use: 0.6331
Component 18:  3 nodes
 - 1 candidates; memory use: 0.6331
Component 19:  3 nodes
 - 1 candidates; memory use: 0.6331
Component 20:  3 nodes
 - 1 candidates; memory use: 0.6331
Component 21:  642 nodes
 - 53291 candidates; memory use: 0.6739
Component 22:  3 nodes
 - 1 candidates; memory use: 0.6741
Component 23:  5 nodes
 - 1 candidates; memory use: 0.6741
Component 24:  3 nodes
 - 1 candidates; memory use: 0.6741
Component 25:  3 nodes
 - 1 candidates; memory use: 0.6741
Component 26:  617 nodes
 - 45310 candidates; memory use: 0.7082
Component 27:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 28:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 29:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 30:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 31:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 32:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 33:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 34:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 35:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 36:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 37:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 38:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 39:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 40:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 41:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 42:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 43:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 44:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 45:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 46:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 47:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 48:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 49:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 50:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 51:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 52:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 53:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 54:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 55:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 56:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 57:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 58:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 59:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 60:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 61:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 62:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 63:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 64:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 65:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 66:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 67:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 68:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 69:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 70:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 71:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 72:  1 nodes
 - 0 candidates; memory use: 0.7082
Component 73:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 74:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 75:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 76:  11 nodes
 - 15 candidates; memory use: 0.7082
Component 77:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 78:  3 nodes
 - 1 candidates; memory use: 0.7082
Component 79:  3 nodes
 - 1 candidates; memory use: 0.7082
memory use: 0.7082
Gathering data for dummies...
memory use: 0.7151
Gathering data for candidates...
 - 508248 candidate(s)
 - Candidate 0
 - Candidate 10000
 - Candidate 20000
 - Candidate 30000
 - Candidate 40000
 - Candidate 50000
 - Candidate 60000
 - Candidate 70000
 - Candidate 80000
 - Candidate 90000
 - Candidate 100000
 - Candidate 110000
 - Candidate 120000
 - Candidate 130000
 - Candidate 140000
 - Candidate 150000
 - Candidate 160000
 - Candidate 170000
 - Candidate 180000
 - Candidate 190000
 - Candidate 200000
 - Candidate 210000
 - Candidate 220000
 - Candidate 230000
 - Candidate 240000
 - Candidate 250000
 - Candidate 260000
 - Candidate 270000
 - Candidate 280000
 - Candidate 290000
 - Candidate 300000
 - Candidate 310000
 - Candidate 320000
 - Candidate 330000
 - Candidate 340000
 - Candidate 350000
 - Candidate 360000
 - Candidate 370000
 - Candidate 380000
 - Candidate 390000
 - Candidate 400000
 - Candidate 410000
 - Candidate 420000
 - Candidate 430000
 - Candidate 440000
 - Candidate 450000
 - Candidate 460000
 - Candidate 470000
 - Candidate 480000
 - Candidate 490000
 - Candidate 500000
Traceback (most recent call last):
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/hairyroots.py", line 579, in <module>
    main()
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/hairyroots.py", line 574, in main
    args.func(args)
  File "/dodrio/scratch/projects/starting_2023_001/bpavie/DIRTmu/hairyroots.py", line 207, in run_pipeline
    good_candidates = np.array(all_candidates)[candidate_filter] #[all_candidates[i] for i in candidate_filter]
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (508248,) + inhomogeneous part.

NB:
I created the python conda env using this yaml file on linux, which seems to install create a correct conda env to run DIRT/mu

name: DIRTmu_env_py3
channels:
  - conda-forge
  - defaults
  - conda-forge/label/cf202003
dependencies:
  - python=3.9.10
  - graph-tool
  - pip
  - scipy=1.8.0
  - scikit-image=0.19.2
  - scikit-learn=1.0.2
  - matplotlib=3.5.1
  - matplotlib-base=3.5.1
  - psutil=5.9.0
  - pandas=1.4.1

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions