add support for multiple labels

caliban_toolbox/reshape_data.py

@@ -32,6 +32,7 @@
 
 import xarray as xr
 
+from caliban_toolbox import settings
 from caliban_toolbox.utils import crop_utils, slice_utils, io_utils
 from caliban_toolbox.utils.crop_utils import compute_crop_indices, crop_helper
 
@@ -91,12 +92,15 @@ def crop_multichannel_data(X_data, y_data, crop_size=None, crop_num=None, overla
     if overlap_frac < 0 or overlap_frac > 1:
         raise ValueError('overlap_frac must be between 0 and 1')
 
-    if list(X_data.dims) != ['fovs', 'stacks', 'crops', 'slices', 'rows', 'cols', 'channels']:
+    if list(X_data.dims) != settings.X_DIMENSION_LABELS:
         raise ValueError('X_data does not have expected dims, found {}'.format(X_data.dims))
 
-    if list(y_data.dims) != ['fovs', 'stacks', 'crops', 'slices', 'rows', 'cols', 'channels']:
+    if list(y_data.dims) != settings.Y_DIMENSION_LABELS:
         raise ValueError('y_data does not have expected dims, found {}'.format(y_data.dims))
 
+    if y_data.shape[-1] != 1:
+        raise ValueError('Only one type of segmentation label can be processed at a time')
+
     # check if testing or running all samples
     if test_parameters:
         X_data, y_data = X_data[:1, ...], y_data[:1, ...]
@@ -141,6 +145,7 @@ def crop_multichannel_data(X_data, y_data, crop_size=None, crop_num=None, overla
     log_data['row_padding'] = int(row_padding)
     log_data['col_padding'] = int(col_padding)
     log_data['num_crops'] = X_data_cropped.shape[2]
+    log_data['label_name'] = y_data.dims[-1]
 
     return X_data_cropped, y_data_cropped, log_data
 
@@ -164,12 +169,11 @@ def create_slice_data(X_data, y_data, slice_stack_len, slice_overlap=0):
         raise ValueError('invalid input data shape, '
                          'expected array of len(7), got {}'.format(X_data.shape))
 
-    if len(y_data.shape) != 7:
-        raise ValueError('invalid input data shape, '
-                         'expected array of len(7), got {}'.format(y_data.shape))
+    if list(X_data.dims) != settings.X_DIMENSION_LABELS:
+        raise ValueError('X_data does not have expected dims, found {}'.format(X_data.dims))
 
-    if slice_stack_len > X_data.shape[1]:
-        raise ValueError('slice size is greater than stack length')
+    if list(y_data.dims) != settings.Y_DIMENSION_LABELS:
+        raise ValueError('y_data does not have expected dims, found {}'.format(y_data.dims))
 
     # compute indices for slices
     stack_len = X_data.shape[1]
@@ -216,13 +220,12 @@ def reconstruct_image_stack(crop_dir, verbose=True):
 
     # labels for each index within a dimension
     _, stack_len, _, _, row_len, col_len, _ = log_data['original_shape']
+    label_name = log_data['label_name']
     coordinate_labels = [log_data['fov_names'], range(stack_len), range(1),
-                         range(1), range(row_len), range(col_len), ['segmentation_label']]
+                         range(1), range(row_len), range(col_len), [label_name]]
 
     # labels for each dimension
-    dimension_labels = ['fovs', 'stacks', 'crops', 'slices', 'rows', 'cols', 'channels']
-
     stitched_xr = xr.DataArray(data=image_stack, coords=coordinate_labels,
-                               dims=dimension_labels)
+                               dims=settings.Y_DIMENSION_LABELS)
 
     stitched_xr.to_netcdf(os.path.join(crop_dir, 'stitched_images.xr'))

caliban_toolbox/reshape_data_test.py

@@ -49,7 +49,7 @@ def test_crop_multichannel_data():
 
     test_y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                                  slice_num=slice_num, row_len=row_len, col_len=col_len,
-                                 chan_len=channel_len)
+                                 chan_len=channel_len, last_dim_name='compartments')
 
     X_data_cropped, y_data_cropped, log_data = \
         reshape_data.crop_multichannel_data(X_data=test_X_data,
@@ -133,7 +133,7 @@ def test_create_slice_data():
 
     y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=num_crops,
                             slice_num=num_slices, row_len=row_len, col_len=col_len,
-                            chan_len=chan_len)
+                            chan_len=chan_len, last_dim_name='compartments')
 
     X_slice, y_slice, slice_indices = reshape_data.create_slice_data(X_data, y_data,
                                                                      slice_stack_len)
@@ -155,7 +155,8 @@ def test_reconstruct_image_stack():
 
         y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                                 slice_num=slice_num,
-                                row_len=row_len, col_len=col_len, chan_len=1)
+                                row_len=row_len, col_len=col_len, chan_len=1,
+                                last_dim_name='compartments')
 
         # create image with artificial objects to be segmented
 
@@ -203,7 +204,8 @@ def test_reconstruct_image_stack():
 
         y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                                 slice_num=slice_num,
-                                row_len=row_len, col_len=col_len, chan_len=1)
+                                row_len=row_len, col_len=col_len, chan_len=1,
+                                last_dim_name='compartments')
 
         # tag upper left hand corner of the label in each image
         tags = np.arange(stack_len)
@@ -237,7 +239,8 @@ def test_reconstruct_image_stack():
 
         y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                                 slice_num=slice_num,
-                                row_len=row_len, col_len=col_len, chan_len=1)
+                                row_len=row_len, col_len=col_len, chan_len=1,
+                                last_dim_name='compartments')
 
         # create image with artificial objects to be segmented
 

caliban_toolbox/settings.py

@@ -0,0 +1,38 @@
+# Copyright 2016-2020 The Van Valen Lab at the California Institute of
+# Technology (Caltech), with support from the Paul Allen Family Foundation,
+# Google, & National Institutes of Health (NIH) under Grant U24CA224309-01.
+# All rights reserved.
+#
+# Licensed under a modified Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.github.com/vanvalenlab/caliban-toolbox/LICENSE
+#
+# The Work provided may be used for non-commercial academic purposes only.
+# For any other use of the Work, including commercial use, please contact:
+# vanvalenlab@gmail.com
+#
+# Neither the name of Caltech nor the names of its contributors may be used
+# to endorse or promote products derived from this software without specific
+# prior written permission.
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+from __future__ import absolute_import
+from __future__ import print_function
+from __future__ import division
+
+from decouple import config
+
+X_DIMENSION_LABELS = config('DIMENSION_LABELS', cast=list,
+                            default=['fovs', 'stacks', 'crops', 'slices',
+                                     'rows', 'cols', 'channels'])
+
+Y_DIMENSION_LABELS = config('DIMENSION_LABELS', cast=list,
+                            default=['fovs', 'stacks', 'crops', 'slices',
+                                     'rows', 'cols', 'compartments'])

caliban_toolbox/utils/crop_utils.py

@@ -86,7 +86,7 @@ def crop_helper(input_data, row_starts, row_ends, col_starts, col_ends, padding)
     """Crops an image into pieces according to supplied coordinates
 
     Args:
-        input_data: xarray of [fovs, stacks, crops, slices, rows, cols, channels] to be cropped
+        input_data: xarray of either X or y data to be cropped
         row_starts: list of indices where row crops start
         row_ends: list of indices where row crops end
         col_starts: list of indices where col crops start
@@ -104,6 +104,9 @@ def crop_helper(input_data, row_starts, row_ends, col_starts, col_ends, padding)
     if input_crop_num > 1:
         raise ValueError("Array has already been cropped")
 
+    # get name of last dimension from input data to determine if X or y
+    last_dim_name = input_data.dims[-1]
+
     crop_num = len(row_starts) * len(col_starts)
     crop_size_row = row_ends[0] - row_starts[0]
     crop_size_col = col_ends[0] - col_starts[0]
@@ -114,12 +117,10 @@ def crop_helper(input_data, row_starts, row_ends, col_starts, col_ends, padding)
 
     # labels for each index within a dimension
     coordinate_labels = [input_data.fovs, input_data.stacks, range(crop_num), input_data.slices,
-                         range(crop_size_row), range(crop_size_col), input_data.channels]
+                         range(crop_size_row), range(crop_size_col), input_data[last_dim_name]]
 
     # labels for each dimension
-    dimension_labels = ['fovs', 'stacks', 'crops', 'slices', 'rows', 'cols', 'channels']
-
-    cropped_xr = xr.DataArray(data=cropped_stack, coords=coordinate_labels, dims=dimension_labels)
+    cropped_xr = xr.DataArray(data=cropped_stack, coords=coordinate_labels, dims=input_data.dims)
 
     # pad the input to account for imperfectly overlapping final crop in rows and cols
     formatted_padding = ((0, 0), (0, 0), (0, 0), (0, 0), (0, padding[0]), (0, padding[1]), (0, 0))

caliban_toolbox/utils/crop_utils_test.py

@@ -32,7 +32,8 @@
 import xarray as xr
 
 
-def _blank_data_xr(fov_len, stack_len, crop_num, slice_num, row_len, col_len, chan_len):
+def _blank_data_xr(fov_len, stack_len, crop_num, slice_num, row_len, col_len, chan_len,
+                   last_dim_name='channels'):
     """Test function to generate a blank xarray with the supplied dimensions
 
     Inputs
@@ -43,6 +44,7 @@ def _blank_data_xr(fov_len, stack_len, crop_num, slice_num, row_len, col_len, ch
         row_num: number of rows
         col_num: number of cols
         chan_num: number of channels
+        last_dim_name: name of last dimension. Either channels or compartments for X or y data
 
     Outputs
         test_xr: xarray of [fov_num, row_num, col_num, chan_num]"""
@@ -56,7 +58,7 @@ def _blank_data_xr(fov_len, stack_len, crop_num, slice_num, row_len, col_len, ch
                                  coords=[fovs, range(stack_len), range(crop_num), range(slice_num),
                                          range(row_len), range(col_len), channels],
                                  dims=["fovs", "stacks", "crops", "slices",
-                                       "rows", "cols", "channels"])
+                                       "rows", "cols", last_dim_name])
 
     return test_stack_xr
 
@@ -195,7 +197,8 @@ def test_stitch_crops():
 
     y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                             slice_num=slice_num,
-                            row_len=row_len, col_len=col_len, chan_len=1)
+                            row_len=row_len, col_len=col_len, chan_len=1,
+                            last_dim_name='compartments')
 
     # create image with artificial objects to be segmented
 
@@ -258,7 +261,8 @@ def test_stitch_crops():
 
     y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                             slice_num=slice_num,
-                            row_len=row_len, col_len=col_len, chan_len=chan_len)
+                            row_len=row_len, col_len=col_len, chan_len=chan_len,
+                            last_dim_name='compartments')
     side_len = 40
     cell_num = y_data.shape[4] // side_len
 
@@ -309,7 +313,7 @@ def test_stitch_crops():
     log_data["num_crops"] = y_cropped.shape[2]
     log_data["original_shape"] = y_data.shape
     log_data["fov_names"] = y_data.fovs.values.tolist()
-    log_data["channel_names"] = y_data.channels.values.tolist()
+    log_data["label_name"] = str(y_data.coords['compartments'][0].values)
 
     stitched_img = crop_utils.stitch_crops(crop_stack=y_cropped, log_data=log_data)
 

caliban_toolbox/utils/io_utils.py

@@ -122,7 +122,7 @@ def save_npzs_for_caliban(X_data, y_data, original_data, log_data, save_dir,
                 raise NotImplementedError()
 
     log_data['fov_names'] = fov_names.tolist()
-    log_data['channel_names'] = original_data.channels.values.tolist()
+    log_data['label_name'] = str(y_data.coords[y_data.dims[-1]][0].values)
     log_data['original_shape'] = original_data.shape
     log_data['slice_stack_len'] = X_data.shape[1]
     log_data['save_format'] = save_format

caliban_toolbox/utils/io_utils_test.py

@@ -47,7 +47,8 @@ def test_save_npzs_for_caliban():
 
     y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=num_crops,
                             slice_num=num_slices,
-                            row_len=row_len, col_len=col_len, chan_len=1)
+                            row_len=row_len, col_len=col_len, chan_len=1,
+                            last_dim_name='compartments')
 
     sliced_X, sliced_y, log_data = reshape_data.create_slice_data(X_data=X_data, y_data=y_data,
                                                                   slice_stack_len=slice_stack_len)
@@ -194,7 +195,8 @@ def test_load_npzs():
 
         y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                                 slice_num=slice_num,
-                                row_len=row_len, col_len=col_len, chan_len=1)
+                                row_len=row_len, col_len=col_len, chan_len=1,
+                                last_dim_name='compartments')
 
         # slice the data
         X_slice, y_slice, log_data = reshape_data.create_slice_data(X_data, y_data,
@@ -249,7 +251,8 @@ def test_load_npzs():
 
         y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                                 slice_num=slice_num,
-                                row_len=row_len, col_len=col_len, chan_len=1)
+                                row_len=row_len, col_len=col_len, chan_len=1,
+                                last_dim_name='compartments')
 
         # slice the data
         X_slice, y_slice, log_data = reshape_data.create_slice_data(X_data,

caliban_toolbox/utils/slice_utils.py

@@ -86,6 +86,9 @@ def slice_helper(data_xr, slice_start_indices, slice_end_indices):
     if input_slice_num > 1:
         raise ValueError('Input array already contains slice data')
 
+    # get name of last dimension from input data to determine if X or y
+    last_dim_name = data_xr.dims[-1]
+
     slice_num = len(slice_start_indices)
     sliced_stack_len = slice_end_indices[0] - slice_start_indices[0]
 
@@ -95,12 +98,10 @@ def slice_helper(data_xr, slice_start_indices, slice_end_indices):
 
     # labels for each index within a dimension
     coordinate_labels = [data_xr.fovs, range(sliced_stack_len), range(crop_num), range(slice_num),
-                         range(row_len), range(col_len), data_xr.channels]
+                         range(row_len), range(col_len), data_xr[last_dim_name]]
 
     # labels for each dimension
-    dimension_labels = ['fovs', 'stacks', 'crops', 'slices', 'rows', 'cols', 'channels']
-
-    slice_xr = xr.DataArray(data=slice_data, coords=coordinate_labels, dims=dimension_labels)
+    slice_xr = xr.DataArray(data=slice_data, coords=coordinate_labels, dims=data_xr.dims)
 
     # loop through slice indices to generate sliced data
     slice_counter = 0

caliban_toolbox/utils/slice_utils_test.py

@@ -148,7 +148,8 @@ def test_stitch_slices():
 
     y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                             slice_num=slice_num,
-                            row_len=row_len, col_len=col_len, chan_len=1)
+                            row_len=row_len, col_len=col_len, chan_len=1,
+                            last_dim_name='compartments')
 
     # generate ordered data
     linear_seq = np.arange(stack_len * row_len * col_len)
@@ -178,7 +179,8 @@ def test_stitch_slices():
 
     y_data = _blank_data_xr(fov_len=fov_len, stack_len=stack_len, crop_num=crop_num,
                             slice_num=slice_num,
-                            row_len=row_len, col_len=col_len, chan_len=1)
+                            row_len=row_len, col_len=col_len, chan_len=1,
+                            last_dim_name='compartments')
 
     # generate ordered data
     linear_seq = np.arange(stack_len * row_len * col_len)

requirements.txt

@@ -12,3 +12,4 @@ xarray==0.13.0
 netCDF4==1.5.3
 pathlib==1.0.1
 deepcell-toolbox>=0.6.1
+python-decouple==3.1