add layer-type option for load-tif

chunkflow/chunk/base.py

@@ -206,7 +206,10 @@ def clone(self):
         )
 
     @classmethod
-    def from_tif(cls, file_name: str, voxel_offset: tuple=None, dtype: str = None,
+    def from_tif(cls, file_name: str, 
+            voxel_offset: tuple=None, 
+            dtype: str = None,
+            layer_type: str = None,
             voxel_size: tuple=None):
         assert os.path.exists(file_name)
         if os.path.isfile(file_name):
@@ -230,7 +233,7 @@ def from_tif(cls, file_name: str, voxel_offset: tuple=None, dtype: str = None,
                 arr[idx+1, :, :] = section
 
         print(f'read tif chunk with size of {arr.shape}, voxel offset: {voxel_offset}, voxel size: {voxel_size}')
-        return cls(arr, voxel_offset=voxel_offset, voxel_size=voxel_size)
+        return cls(arr, voxel_offset=voxel_offset, voxel_size=voxel_size, layer_type=layer_type)
 
     def to_tif(self, file_name: str=None, compression: str = 'zlib'):
         if file_name is None:

chunkflow/flow/flow.py

@@ -925,14 +925,17 @@ def load_png(tasks: dict, path: str,
               help='global offset of this chunk')
 @click.option('--voxel-size', '-s', type=click.INT, nargs=3, default=None, callback=default_none,
               help='physical size of voxels. The unit is assumed to be nm.')
+@click.option('--layer-type', '-l',
+    type=click.Choice(['image', 'segmentation']), default=None,
+    help='the layer type in neuroglancer for visualization.')
 @click.option('--dtype', '-d',
               type=click.Choice(['uint8', 'uint16', 'uint32', 'uint64', 'float32', 'float64', 'float16']),
               help='convert to data type')
 @click.option('--output-chunk-name', '-o', type=str, default='chunk',
               help='chunk name in the global state')
 @operator
 def load_tif(tasks, name: str, file_name: str, voxel_offset: tuple,
-             voxel_size: tuple, dtype: str, output_chunk_name: str):
+             voxel_size: tuple, layer_type: str, dtype: str, output_chunk_name: str):
     """Read tiff files."""
     for task in tasks:
         if task is not None:
@@ -941,6 +944,7 @@ def load_tif(tasks, name: str, file_name: str, voxel_offset: tuple,
                 file_name,
                 dtype=dtype,
                 voxel_offset=voxel_offset,
+                layer_type=layer_type,
                 voxel_size=voxel_size)
             task['log']['timer'][name] = time() - start
         yield task

chunkflow/flow/neuroglancer.py

@@ -373,7 +373,7 @@ def parse_selected_args(varname: str) -> Tuple[str, dict]:
                 name, layer_kwargs = parse_selected_args(name)
                 data = datas[name]
                 layer_args = (viewer_state, name, data)
-                # breakpoint()
+                breakpoint()
 
                 if data is None:
                     continue
@@ -401,7 +401,7 @@ def parse_selected_args(varname: str) -> Tuple[str, dict]:
                             raise ValueError('affinity map is not working yet. To-Do.')
                         else:
                             raise ValueError('unsupported data type.')
-                    if data.layer_type == 'segmentation':
+                    elif data.layer_type == 'segmentation':
                         self._append_segmentation_layer(*layer_args, **layer_kwargs)
                     elif data.layer_type == 'probability_map':
                         self._append_probability_map_layer(*layer_args, **layer_kwargs)

chunkflow/plugins/agglomerate.py

@@ -5,14 +5,14 @@
 from waterz import agglomerate
 
 
-threshold: float = 0.7
-aff_threshold_low: float = 0.001
-aff_threshold_high: float = 0.9999
-scoring_function: str = 'OneMinus<MeanAffinity<RegionGraphType, ScoreValue>>'
-flip_channel: bool = True
-
-
-def execute(affs: Chunk, fragments: np.ndarray = None):
+def execute(affs: Chunk, 
+        fragments: np.ndarray = None, 
+        threshold: float = 0.7,
+        aff_threshold_low: float = 0.001,
+        aff_threshold_high: float = 0.9999,
+        scoring_function: str = 'OneMinus<MeanAffinity<RegionGraphType, ScoreValue>>',
+        flip_channel: bool = True,
+    ):
     """
     Mean/max agglomeration of affinity map including watershed step.