Merge pull request open-mmlab#197 from open-mmlab/zz/typo

Typos

mmedit/core/hooks/visualization.py

@@ -24,7 +24,7 @@ class VisualizationHook(Hook):
         filename_tmpl (str): Format string used to save images. The output file
             name will be formatted as this args. Default: 'iter_{}.png'.
         rerange (bool): Whether to rerange the output value from [-1, 1] to
-            [0, 1]. We highly recommand users should preprocess the
+            [0, 1]. We highly recommend users should preprocess the
             visualization results on their own. Here, we just provide a simple
             interface. Default: True.
         bgr2rgb (bool): Whether to reformat the channel dimension from BGR to

mmedit/models/backbones/encoder_decoders/decoders/indexnet_decoder.py

@@ -53,7 +53,7 @@ def forward(self, x, shortcut, dec_idx_feat=None):
 
         Args:
             x (Tensor): Input feature map with shape (N, C, H, W).
-            shortcut (Tensor): The shorcut connection with shape
+            shortcut (Tensor): The shortcut connection with shape
                 (N, C, H', W').
             dec_idx_feat (Tensor, optional): The decode index feature map with
                 shape (N, C, H', W'). Defaults to None.
@@ -117,7 +117,7 @@ def init_weights(self):
                 normal_init(m, mean=0, std=std)
 
     def forward(self, inputs):
-        """Forward fucntion.
+        """Forward function.
 
         Args:
             inputs (dict): Output dict of IndexNetEncoder.

mmedit/models/backbones/encoder_decoders/decoders/pconv_decoder.py

@@ -10,12 +10,12 @@
 class PConvDecoder(nn.Module):
     """Decoder with partial conv.
 
-    About the details for this archetecture, pls see:
+    About the details for this architecture, pls see:
     Image Inpainting for Irregular Holes Using Partial Convolutions
 
     Args:
         num_layers (int): The number of convolutional layers. Default: 7.
-        interpolation (str): The upsamle mode. Default: 'nearest'.
+        interpolation (str): The upsample mode. Default: 'nearest'.
         conv_cfg (dict): Config for convolution module. Default:
             {'type': 'PConv', 'multi_channel': True}.
         norm_cfg (dict): Config for norm layer. Default:

mmedit/models/backbones/encoder_decoders/encoders/pconv_encoder.py

@@ -9,7 +9,7 @@
 class PConvEncoder(nn.Module):
     """Encoder with partial conv.
 
-    About the details for this archetecture, pls see:
+    About the details for this architecture, pls see:
     Image Inpainting for Irregular Holes Using Partial Convolutions
 
     Args:

mmedit/models/common/aspp.py

@@ -114,7 +114,7 @@ def forward(self, x):
             x (Tensor): Input tensor with shape (N, C, H, W).
 
         Returns:
-            Tensor: Outpur tensor.
+            Tensor: Output tensor.
         """
         res = []
         for conv in self.convs:

mmedit/models/common/separable_conv_module.py

@@ -89,7 +89,7 @@ def forward(self, x):
             x (Tensor): Input tensor with shape (N, C, H, W).
 
         Returns:
-            Tensor: Outpur tensor.
+            Tensor: Output tensor.
         """
         x = self.depthwise_conv(x)
         x = self.pointwise_conv(x)

mmedit/models/components/discriminators/multi_layer_disc.py

@@ -15,7 +15,7 @@ class MultiLayerDiscriminator(nn.Module):
 
     Args:
         in_channels (int): Input channel of the first input convolution.
-        max_channels (int): The maxinum channel number in this structure.
+        max_channels (int): The maximum channel number in this structure.
         num_conv (int): Number of stacked intermediate convs (including input
             conv but excluding output conv).
         fc_in_channels (int | None): Input dimension of the fully connected

mmedit/models/components/refiners/deepfill_refiner.py

@@ -62,7 +62,7 @@ def forward(self, x, mask):
         # contextual attention branch
         attention_x = self.encoder_attention(x)['out']
         h_x, w_x = attention_x.shape[-2:]
-        # resale mask to a samller size
+        # resale mask to a smaller size
         resized_mask = F.interpolate(mask, size=(h_x, w_x))
         attention_x, offset = self.contextual_attention_neck(
             attention_x, resized_mask)

mmedit/models/losses/perceptual_loss.py

@@ -19,7 +19,7 @@ class PerceptualVGG(nn.Module):
             forward function will return the corresponding features. This
             list contains the name each layer in `vgg.feature`. An example
             of this list is ['4', '10'].
-        vgg_tyep (str): Set the type of vgg network. Default: 'vgg19'.
+        vgg_type (str): Set the type of vgg network. Default: 'vgg19'.
         use_input_norm (bool): If True, normalize the input image.
             Importantly, the input feature must in the range [0, 1].
             Default: True.
@@ -58,7 +58,7 @@ def __init__(self,
                 torch.Tensor([0.229, 0.224, 0.225]).view(1, 3, 1, 1))
 
         for v in self.vgg_layers.parameters():
-            v.requies_grad = False
+            v.requires_grad = False
 
     def forward(self, x):
         """Forward function.
@@ -105,14 +105,14 @@ class PerceptualLoss(nn.Module):
         use_input_norm (bool):  If True, normalize the input image in vgg.
             Default: True.
         perceptual_weight (float): If `perceptual_weight > 0`, the perceptual
-            loss will be calculated and the loss will multiplified by the
+            loss will be calculated and the loss will multiplied by the
             weight. Default: 1.0.
-        style_weight (flaot): If `style_weight > 0`, the style loss will be
-            calculated and the loss will multiplified by the weight.
+        style_weight (float): If `style_weight > 0`, the style loss will be
+            calculated and the loss will multiplied by the weight.
             Default: 1.0.
         norm_img (bool): If True, the image will be normed to [0, 1]. Note that
             this is different from the `use_input_norm` which norm the input in
-            in forward fucntion of vgg according to the statistics of dataset.
+            in forward function of vgg according to the statistics of dataset.
             Importantly, the input image must be in range [-1, 1].
         pretrained (str): Path for pretrained weights. Default:
             'torchvision://vgg19'
@@ -163,7 +163,7 @@ def forward(self, x, gt):
         x_features = self.vgg(x)
         gt_features = self.vgg(gt.detach())
 
-        # calculate preceptual loss
+        # calculate perceptual loss
         if self.perceptual_weight > 0:
             percep_loss = 0
             for k in x_features.keys():

mmedit/utils/logger.py

@@ -9,7 +9,7 @@ def get_root_logger(log_file=None, log_level=logging.INFO):
     The logger will be initialized if it has not been initialized. By default a
     StreamHandler will be added. If `log_file` is specified, a FileHandler will
     also be added. The name of the root logger is the top-level package name,
-    e.g., "mmsr".
+    e.g., "mmedit".
 
     Args:
         log_file (str | None): The log filename. If specified, a FileHandler

tests/test_crop.py

@@ -315,7 +315,7 @@ def test_crop_around_fg(self):
             # keys must contain 'seg'
             CropAroundFg(['fg', 'bg'])
         with pytest.raises(TypeError):
-            # bd_ratio_range must be a tuple of 2 flaot
+            # bd_ratio_range must be a tuple of 2 float
             CropAroundFg(['seg', 'merged'], bd_ratio_range=0.1)
 
         keys = ['bg', 'merged', 'seg']

tools/evaluate_comp1k.py

@@ -35,7 +35,7 @@ def evaluate_one(args):
 def evaluate(pred_root, gt_root, trimap_root, verbose, nproc):
     """Evaluate test results of Adobe composition-1k dataset.
 
-    There are 50 different ground truth foregrounds and alpha mattes paire,
+    There are 50 different ground truth foregrounds and alpha mattes pairs,
     each of the foreground will be composited with 20 different backgrounds,
     producing 1000 images for testing. In some repo, the ground truth alpha
     matte will be copied 20 times and named the same as the images. This