PMD Compression

Simple PMD compression, the hello world of masknmf

/home/runner/work/masknmf-toolbox/masknmf-toolbox/masknmf/utils/_cuda.py:7: UserWarning: You've explicitly selected to perform computations on the cpu, performance will be significantly slower
  warn(
[25-12-03 05:22:14]: Starting compression
[25-12-03 05:22:14]: sampled from the following regions: [0]
[25-12-03 05:22:14]: We are initializing on a total of 2000 frames
[25-12-03 05:22:14]: Loading data to estimate complete spatial basis
[25-12-03 05:22:14]: skipping the pruning step for frame cutoff
[25-12-03 05:22:14]: Finding spatiotemporal roughness thresholds

  0%|          | 0/250 [00:00<?, ?it/s]
  3%|▎         | 7/250 [00:00<00:03, 69.61it/s]
  6%|▌         | 15/250 [00:00<00:03, 71.52it/s]
  9%|▉         | 23/250 [00:00<00:03, 73.06it/s]
 12%|█▏        | 31/250 [00:00<00:02, 73.47it/s]
 16%|█▌        | 39/250 [00:00<00:02, 73.39it/s]
 19%|█▉        | 47/250 [00:00<00:02, 73.38it/s]
 22%|██▏       | 55/250 [00:00<00:02, 73.35it/s]
 25%|██▌       | 63/250 [00:00<00:02, 72.64it/s]
 28%|██▊       | 71/250 [00:00<00:02, 72.81it/s]
 32%|███▏      | 79/250 [00:01<00:02, 73.30it/s]
 35%|███▍      | 87/250 [00:01<00:02, 73.39it/s]
 38%|███▊      | 95/250 [00:01<00:02, 73.56it/s]
 41%|████      | 103/250 [00:01<00:01, 73.85it/s]
 44%|████▍     | 111/250 [00:01<00:01, 71.57it/s]
 48%|████▊     | 119/250 [00:01<00:01, 72.41it/s]
 51%|█████     | 127/250 [00:01<00:01, 73.00it/s]
 54%|█████▍    | 135/250 [00:01<00:01, 73.12it/s]
 57%|█████▋    | 143/250 [00:01<00:01, 73.37it/s]
 60%|██████    | 151/250 [00:02<00:01, 73.24it/s]
 64%|██████▎   | 159/250 [00:02<00:01, 73.11it/s]
 67%|██████▋   | 167/250 [00:02<00:01, 73.31it/s]
 70%|███████   | 175/250 [00:02<00:01, 73.62it/s]
 73%|███████▎  | 183/250 [00:02<00:00, 73.95it/s]
 76%|███████▋  | 191/250 [00:02<00:00, 74.20it/s]
 80%|███████▉  | 199/250 [00:02<00:00, 74.61it/s]
 83%|████████▎ | 207/250 [00:02<00:00, 74.23it/s]
 86%|████████▌ | 215/250 [00:02<00:00, 74.22it/s]
 89%|████████▉ | 223/250 [00:03<00:00, 74.17it/s]
 92%|█████████▏| 231/250 [00:03<00:00, 74.24it/s]
 96%|█████████▌| 239/250 [00:03<00:00, 74.26it/s]
 99%|█████████▉| 247/250 [00:03<00:00, 74.02it/s]
100%|██████████| 250/250 [00:03<00:00, 73.46it/s]
[25-12-03 05:22:18]: Running Blockwise Decompositions
[25-12-03 05:22:18]: Constructed U matrix. Rank of U is 184
[25-12-03 05:22:18]: PMD Objected constructed
/opt/hostedtoolcache/Python/3.12.12/x64/lib/python3.12/site-packages/fastplotlib/graphics/features/_base.py:18: UserWarning: casting float64 array to float32
  warn(f"casting {array.dtype} array to float32")

# test_example = true

import masknmf
import torch
import fastplotlib as fpl
from urllib.request import urlretrieve
import tifffile


urlretrieve(
    "https://github.com/flatironinstitute/CaImAn/raw/refs/heads/main/example_movies/demoMovie.tif",
    "./demo.tif"
)

# always lazy load raw data by memmaping or other methods
data = tifffile.imread("./demo.tif")

block_sizes = [32, 32]
max_components = 20

# it's recommended to use masknmf on a machine with a GPU
# device = "cuda" if torch.cuda.is_available() else "cpu"
device = "cpu"
# number of frames used to estimate the spatial basis in PMD
num_frames_for_spatial_fit = data.shape[0]

# perform PMD
pmd_result = masknmf.compression.pmd_decomposition(
    data,
    block_sizes,
    num_frames_for_spatial_fit,
    max_components=max_components,
    device=device,
    frame_batch_size=1024
)

# get the residual
pmd_residual = masknmf.PMDResidualArray(data, pmd_result)

# view the movies, note that all these array are LAZY evaluated, allowing you to view extremely large datasets!
iw = fpl.ImageWidget(
    data=[data, pmd_result, pmd_residual],
    names=["raw", "pmd", "residual"],
    figure_kwargs={"size": (1000, 340), "shape": (1, 3)},
    cmap="gnuplot2",
)

iw.show()

# use the time slider or set the frame index programmatically
iw.current_index = {"t": 1610}

# manually set vmin-vmax to emphasize noise in raw video
# you can also adjust the vmin-vmax using the histogram tool
# reset the vmin-vmax by clicking the buttons under "ImageWidget Controls"
for image in iw.managed_graphics:
    image.vmax = 3_200

# remove toolbar to reduce clutter
for subplot in iw.figure:
    subplot.toolbar = False


# ignore the remaining lines these are just for docs generation
figure = iw.figure
if __name__ == "__main__":
    print(__doc__)
    fpl.loop.run()