fiberpipeline/config.yaml at main · linok-bc/fiberpipeline · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
dataset_path: "/projects/weilab/dataset/barcode/LZ58_test"
output_path: "outputs"
anisotropy: [40, 16.3, 16.3]
to_um: 0.01

num_cpus: 48
num_gpus: 1

fiber_segmentation_model:
  dataset_path: "/projects/weilab/dataset/barcode/train"

fiber_segmentation:
  batch_size: 64

  _path: fiberpipeline/segmentation/fiber_segmentation_model
  checkpoint: ${fiber_segmentation._path}/checkpoints/checkpoint_100000.pth.tar
  base_yaml: ${fiber_segmentation._path}/configs/base_config.yaml
  bcs_yaml: ${fiber_segmentation._path}/configs/bcs_config.yaml

  clahe:
    tile_size: 8
    clip_limit: 2.0

  bcs:
    thres1: 0.21290304093127832
    thres2: 0.6229526243971144
    thres3: 0
    thres4: 0
    thres5: 0
    thres_small: 10


cell_segmentation:
  model_type: "vit_b_lm" # base (also t for tiny; l for large), light microscopy
  visualize: true

skeletonize_fibers:
  manual_z_scale: 0.33
  percentile_fit: [0.1, 0.9]
  num_centerline_points: 1000
  spline_smoothing: null
  extrapolate: [-0.2, 1.2]

normalize_signals:
  gaussian_filter_ratio: 0.01 # for cropping
  midpoint_range: [0.45, 0.55] # for determining midpoint
  scale_range: [-0.2, 0.2] # if negative, scale left up, if positive, scale right up
  num_trials: 100 # number of bayesian trials to find the best midpoint and scaling

validate_fibers:
  thres_length: 800
  thres_pca_ratio: 0.0
  thres_mean_soma: 0.0
  one_per_soma: false

stats:
  length_thresh: [800, 2500]

precomputed:
  jobs: 64
  chunk_size: [16, 64, 64]
  downsample_factors:
    - [1, 2, 2]
    - [2, 2, 2]
    - [2, 2, 2]
  mesh_mip: 2
  mesh_merge_magnitude: 3
  im_shader: |
    #uicontrol float brightness slider(min=-10, max=10, default=0, step=0.1)
    #uicontrol invlerp normalized
    #uicontrol vec3 color color
    void main() {
      float val = normalized();
      float alpha = smoothstep(0.0, 1.0, val);
      vec3 adjustedColor = color * exp(brightness);
      emitRGBA(vec4(adjustedColor, alpha));
    }

napari:
  # skel_downsample: 10
  skel_downsample: 5