fixing training bug (#867)

cellpose/__main__.py

@@ -231,7 +231,8 @@ def main():
                     model.net, images, labels, train_files=image_names,
                     test_data=test_images, test_labels=test_labels,
                     test_files=image_names_test, learning_rate=args.learning_rate,
-                    weight_decay=args.weight_decay, channels=channels,
+                    weight_decay=args.weight_decay, channels=channels, 
+                    channel_axis=args.channel_axis,
                     save_path=os.path.realpath(args.dir), save_every=args.save_every,
                     SGD=args.SGD, n_epochs=args.n_epochs, batch_size=args.batch_size,
                     min_train_masks=args.min_train_masks,

cellpose/models.py

@@ -283,9 +283,8 @@ def __init__(self, gpu=False, pretrained_model=False, model_type=None,
         self.nchan = nchan
         self.nclasses = 3
         nbase = [32, 64, 128, 256]
-        self.nchan = nchan
         self.nbase = [nchan, *nbase]
-
+        
         self.net = CPnet(self.nbase, self.nclasses, sz=3, mkldnn=self.mkldnn,
                          max_pool=True, diam_mean=diam_mean).to(self.device)
 

cellpose/train.py

@@ -60,7 +60,7 @@ def _get_batch(inds, data=None, labels=None, files=None, labels_files=None,
         if channels is not None:
             imgs = [
                 transforms.convert_image(img, channels=channels,
-                                         channel_axis=channel_axis) for img in imgs
+                                         channel_axis=channel_axis, nchan=None) for img in imgs
             ]
             imgs = [img.transpose(2, 0, 1) for img in imgs]
         if normalize_params["normalize"]:
@@ -89,9 +89,9 @@ def _reshape_norm(data, channels=None, channel_axis=None,
     Returns:
         list: List of reshaped and normalized data.
     """
-    if channels is not None:
+    if channels is not None or channel_axis is not None:
         data = [
-            transforms.convert_image(td, channels=channels, channel_axis=channel_axis)
+            transforms.convert_image(td, channels=channels, channel_axis=channel_axis, nchan=None)
             for td in data
         ]
         data = [td.transpose(2, 0, 1) for td in data]
@@ -111,7 +111,7 @@ def _reshape_norm_save(files, channels=None, channel_axis=None,
         td = io.imread(f)
         if channels is not None:
             td = transforms.convert_image(td, channels=channels,
-                                          channel_axis=channel_axis)
+                                          channel_axis=channel_axis, nchan=None)
             td = td.transpose(2, 0, 1)
         if normalize_params["normalize"]:
             td = transforms.normalize_img(td, normalize=normalize_params, axis=0)
@@ -336,7 +336,6 @@ def train_seg(net, train_data=None, train_labels=None, train_files=None,
     Returns:
         Path: path to saved model weights
     """
-
     device = net.device
 
     scale_range = 0.5 if rescale else 1.0

cellpose/transforms.py

@@ -514,7 +514,7 @@ def convert_image(x, channels, channel_axis=None, z_axis=None, do_3D=False, ncha
 
     else:
         # code above put channels last
-        if x.shape[-1] > nchan:
+        if nchan is not None and x.shape[-1] > nchan:
             transforms_logger.warning(
                 "WARNING: more than %d channels given, use 'channels' input for specifying channels - just using first %d channels to run processing"
                 % (nchan, nchan))
@@ -524,7 +524,7 @@ def convert_image(x, channels, channel_axis=None, z_axis=None, do_3D=False, ncha
             transforms_logger.critical("ERROR: cannot process 4D images in 2D mode")
             raise ValueError("ERROR: cannot process 4D images in 2D mode")
 
-        if x.shape[-1] < nchan:
+        if nchan is not None and x.shape[-1] < nchan:
             x = np.concatenate((x, np.tile(np.zeros_like(x), (1, 1, nchan - 1))),
                                axis=-1)