Remove install dep on torchvision

CHANGELOG.md

@@ -22,6 +22,7 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 ### Deprecated
 ### Removed
 - Removed the fluent decorator, just use return self
+- Removed install dependency on `torchvision`, still required for some functionality
 ### Fixed
 - Fixed bug where replay errored when train or val steps were None
 - Fixed a bug where mock optimser wouldn't call it's closure

requirements.txt

@@ -1,5 +1,4 @@
 torch>=0.4
-torchvision<0.3
 numpy
 scikit-learn
 tqdm
@@ -8,4 +7,5 @@ visdom
 livelossplot
 mock
 Pillow
-matplotlib
+matplotlib
+torchvision<0.3

setup.py

@@ -24,6 +24,6 @@
     description='A model training and variational auto-encoder library for pytorch',
     long_description=long_description,
     long_description_content_type='text/markdown',
-    install_requires=['torch>=0.4', 'torchvision<0.3', 'tqdm'],
+    install_requires=['torch>=0.4', 'tqdm'],
     python_requires='>=2.7,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*',
 )

torchbearer/callbacks/tensor_board.py

@@ -3,7 +3,6 @@
 
 import torch
 import torch.nn.functional as F
-import torchvision.utils as utils
 
 import torchbearer
 from torchbearer.callbacks import Callback
@@ -353,7 +352,7 @@ def on_end(self, state):
 
 class TensorBoardImages(AbstractTensorBoard):
     """The TensorBoardImages callback will write a selection of images from the validation pass to tensorboard using the
-    TensorboardX library and torchvision.utils.make_grid. Images are selected from the given key and saved to the given
+    TensorboardX library and torchvision.utils.make_grid (requires torchvision). Images are selected from the given key and saved to the given
     path. Full name of image sub directory will be model name + _ + comment.
 
     Args:
@@ -404,6 +403,7 @@ def __init__(self, log_dir='./logs',
 
     def on_step_validation(self, state):
         if not self.done:
+            import torchvision.utils as utils
             data = state[self.key].clone()
 
             if len(data.size()) == 3:

torchbearer/variational/datasets.py

@@ -5,10 +5,10 @@
 import numpy as np
 from PIL import Image
 from torch.utils.data import Dataset
-from torchvision.datasets.folder import has_file_allowed_extension, default_loader, IMG_EXTENSIONS
 
 
 def make_dataset(dir, extensions):
+    from torchvision.datasets.folder import has_file_allowed_extension
     images = []
     for root, _, fnames in sorted(os.walk(dir)):
         for fname in sorted(fnames):
@@ -20,7 +20,7 @@ def make_dataset(dir, extensions):
 
 
 class SimpleImageFolder(Dataset):
-    def __init__(self, root, loader=default_loader, extensions=IMG_EXTENSIONS, transform=None, target_transform=None):
+    def __init__(self, root, loader=None, extensions=None, transform=None, target_transform=None):
         """
         Simple image folder dataset that loads all images from inside a folder and returns items in (image, image) tuple
 
@@ -31,6 +31,10 @@ def __init__(self, root, loader=default_loader, extensions=IMG_EXTENSIONS, trans
             transform (``Transform``, optional): A function/transform that takes in an PIL image and returns a transformed version. E.g, ``transforms.RandomCrop``
             target_transform (``Transform``, optional): A function/transform that takes in an PIL image and returns a transformed version. E.g, ``transforms.RandomCrop``
         """
+        from torchvision.datasets.folder import default_loader, IMG_EXTENSIONS
+        loader = default_loader if loader is None else loader
+        extensions = IMG_EXTENSIONS if extensions is None else extensions
+
         samples = make_dataset(root, extensions)
 
         self.root = root
@@ -91,6 +95,7 @@ def __init__(self, root, as_npy=False, transform=None):
             as_npy (bool, optional): If True, assume images are stored in numpy arrays. Else assume a standard image format
             transform (``Transform``, optional): A function/transform that takes in an PIL image and returns a transformed version. E.g, ``transforms.RandomCrop``
         """
+        from torchvision.datasets.folder import default_loader, IMG_EXTENSIONS
         if as_npy:
             loader = self.npy_loader
             extensions = ['npy']

torchbearer/variational/visualisation.py

@@ -1,5 +1,4 @@
 import torch
-from torchvision.utils import save_image
 
 import torchbearer as tb
 import torchbearer.callbacks as c
@@ -110,6 +109,7 @@ def vis(self, state):
         raise NotImplementedError
 
     def _save_walk(self, tensor):
+        from torchvision.utils import save_image
         save_image(tensor, self.file, self.row_size, normalize=True, pad_value=1)