refine_flowvision

docs/source/flowvision.models.rst

@@ -6,7 +6,7 @@ Pretrain Models for Visual Tasks
 Classification
 ==============
 
-The models subpackage contains definitions for the following model
+The models' subpackage contains definitions for the following model
 architectures for image classification:
 
 -  `AlexNet`_

docs/source/tutorials/Getting Started.md

@@ -1,7 +1,7 @@
 # Getting Started
 
 ## Installation
-- To install latest stable release of flowvision:
+- To install the latest stable release of flowvision:
 ```bash
 pip install flowvision==0.1.0
 ```
@@ -14,7 +14,7 @@ pip install -e .
 
 ## Usage
 ### Create a model
-In flowvision we support two ways to create a model.
+Flowvision supports two ways to create a model:
 
 - Import the target model from `flowvision.models`, e.g., create `alexnet` from flowvision
 
@@ -77,7 +77,7 @@ You can get the results like:
 ╘════════════════════════════════════════════╧══════════════╛
 ```
 
-### Search for supported model by Wildcard
+### Search for the supported model by Wildcard
 It is easy to search for model architectures by using Wildcard as below:
 ```python
 from flowvision.models import ModelCreator
@@ -108,7 +108,7 @@ You can get the results like:
 ```
 
 ### List all models supported in flowvision
-`ModelCreator.model_list` has similar function as `ModelCreator.model_table` but return a list object, which gives the user a more flexible way to check the supported model in flowvision.
+`ModelCreator.model_list` has a similar function as `ModelCreator.model_table` but returns a list object, which gives the user a more flexible way to check the supported model in flowvision.
 - List all models with pretrained weights
 ```python
 from flowvision.models import ModelCreator

flowvision/data/auto_augment.py

@@ -543,7 +543,7 @@ def auto_augment_transform(config_str, hparams):
 
     :param config_str: String defining configuration of auto augmentation. Consists of multiple sections separated by
         dashes ('-'). The first section defines the AutoAugment policy (one of 'v0', 'v0r', 'original', 'originalr').
-        The remaining sections, not order sepecific determine
+        The remaining sections, not order specific determine
             * 'mstd' -  float std deviation of magnitude noise applied
 
         Example: 'original-mstd0.5' results in AutoAugment with original policy, magnitude_std 0.5
@@ -685,11 +685,11 @@ def rand_augment_transform(config_str, hparams):
 
     :param config_str: String defining configuration of random augmentation. Consists of multiple sections separated by
         dashes ('-'). The first section defines the specific variant of rand augment (currently only 'rand'). The remaining
-        sections, not order sepecific determine
+        sections, not order specific determine
 
             * 'm' - integer magnitude of rand augment
             * 'n' - integer num layers (number of transform ops selected per image)
-            * 'w' - integer probabiliy weight index (index of a set of weights to influence choice of op)
+            * 'w' - integer probability weight index (index of a set of weights to influence choice of op)
             * 'mstd' -  float std deviation of magnitude noise applied, or uniform sampling if infinity (or > 100)
             * 'mmax' - set upper bound for magnitude to something other than default of  _LEVEL_DENOM (10)
             * 'inc' - integer (bool), use augmentations that increase in severity with magnitude (default: 0)
@@ -850,7 +850,7 @@ def augment_and_mix_transform(config_str, hparams):
 
     :param config_str: String defining configuration of random augmentation. Consists of multiple sections separated by
         dashes ('-'). The first section defines the specific variant of rand augment (currently only 'rand'). The remaining
-        sections, not order sepecific determine
+        sections, not order specific determine
 
             * 'm' - integer magnitude (severity) of augmentation mix (default: 3)
             * 'w' - integer width of augmentation chain (default: 3)

flowvision/data/mixup.py

@@ -39,7 +39,7 @@ def mixup_target(target, num_classes, lam=1.0, smoothing=0.0, device="cuda"):
 
 def rand_bbox(img_shape, lam, margin=0.0, count=None):
     """ Standard CutMix bounding-box
-    Generates a random square bbox based on lambda value. This impl includes
+    generates a random square bbox based on lambda value. This impl includes
     support for enforcing a border margin as percent of bbox dimensions.
 
     Args:

flowvision/data/random_erasing.py

@@ -26,7 +26,7 @@ class RandomErasing:
     or single image tensor after it has been normalized by dataset mean and std.
 
     Args:
-        probability: Probability that the Random Erasing operation will be performed
+        probability: Probability that the RandomErasing operation will be performed
         min_area: Minimum percentage of erased area wrt input image area
         max_area: Maximum percentage of erased area wrt input image area
         min_aspect: Minimum aspect ratio of erased area
@@ -36,7 +36,7 @@ class RandomErasing:
             * 'rand'  - erase block is same per-channel random (normal) color
             * 'pixel' - erase block is per-pixel random (normal) color
 
-        max_count: Maximum number of erasing blocks per image, area per box is scaled by count.
+        max_count: Maximum number of erasing blocks per image, and area per box is scaled by count.
             per-image count is randomly chosen between 1 and this value
     """
 

flowvision/models/detection/faster_rcnn.py

@@ -47,7 +47,7 @@ class FasterRCNN(GeneralizedRCNN):
     The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each
     image, and should be in 0-1 range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:
@@ -317,7 +317,7 @@ def fasterrcnn_resnet50_fpn(
     The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each
     images, and should be in ``0-1`` range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:

flowvision/models/detection/fcos.py

@@ -298,7 +298,7 @@ class FCOS(nn.Module):
     The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each
     image, and should be in 0-1 range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:
@@ -729,7 +729,7 @@ def fcos_resnet50_fpn(
     The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each
     image, and should be in ``0-1`` range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:

flowvision/models/detection/mask_rcnn.py

@@ -24,7 +24,7 @@ class MaskRCNN(FasterRCNN):
     The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each
     image, and should be in 0-1 range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:
@@ -275,7 +275,7 @@ def maskrcnn_resnet50_fpn(
     The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each
     image, and should be in ``0-1`` range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:
 

flowvision/models/detection/retinanet.py

@@ -278,7 +278,7 @@ class RetinaNet(nn.Module):
     The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each
     image, and should be in 0-1 range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:
@@ -670,7 +670,7 @@ def retinanet_resnet50_fpn(
     The input to the model is expected to be a list of tensors, each of shape ``[C, H, W]``, one for each
     image, and should be in ``0-1`` range. Different images can have different sizes.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:

flowvision/models/detection/ssd.py

@@ -117,7 +117,7 @@ class SSD(nn.Module):
     image, and should be in 0-1 range. Different images can have different sizes but they will be resized
     to a fixed size before passing it to the backbone.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:
@@ -604,10 +604,10 @@ def ssd300_vgg16(
     Reference: `"SSD: Single Shot MultiBox Detector" <https://arxiv.org/abs/1512.02325>`_.
 
     The input to the model is expected to be a list of tensors, each of shape [C, H, W], one for each
-    image, and should be in 0-1 range. Different images can have different sizes but they will be resized
+    image, and should be in 0-1 range. Different images can have different sizes, but they will be resized
     to a fixed size before passing it to the backbone.
 
-    The behavior of the model changes depending if it is in training or evaluation mode.
+    The behavior of the model changes depending on if it is in training or evaluation mode.
 
     During training, the model expects both the input tensors, as well as a targets (list of dictionary),
     containing:

flowvision/transforms/functional.py

@@ -988,7 +988,7 @@ def rotate(
 
 def rgb_to_grayscale(img: Tensor, num_output_channels: int = 1) -> Tensor:
     """Convert RGB image to grayscale version of image.
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions
 
     Note:

flowvision/transforms/transforms.py

@@ -507,7 +507,7 @@ def __call__(self, img):
 
 
 class RandomChoice(RandomTransforms):
-    """Apply single transformation randomly picked from a list.
+    """Apply a single transformation randomly picked from a list.
     """
 
     def __call__(self, img):
@@ -635,7 +635,7 @@ def __repr__(self):
 
 class RandomHorizontalFlip(Module):
     """Horizontally flip the given image randomly with a given probability.
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading
     dimensions
 
@@ -666,7 +666,7 @@ def __repr__(self):
 
 class RandomVerticalFlip(Module):
     """Vertically flip the given image randomly with a given probability.
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading
     dimensions
 
@@ -698,7 +698,7 @@ def __repr__(self):
 class RandomResizedCrop(Module):
     """Crop a random portion of image and resize it to a given size.
 
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions
 
     A crop of the original image is made: the crop has a random area (H * W)
@@ -833,7 +833,7 @@ def __init__(self, *args, **kwargs):
 
 class FiveCrop(Module):
     """Crop the given image into four corners and the central crop.
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading
     dimensions
 
@@ -882,7 +882,7 @@ def __repr__(self):
 class TenCrop(Module):
     """Crop the given image into four corners and the central crop plus the flipped version of
     these (horizontal flipping is used by default).
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading
     dimensions
 
@@ -934,7 +934,7 @@ def __repr__(self):
 
 class ColorJitter(Module):
     """Randomly change the brightness, contrast, saturation and hue of an image.
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions.
     If img is PIL Image, mode "1", "L", "I", "F" and modes with transparency (alpha channel) are not supported.
 
@@ -1078,7 +1078,7 @@ def __repr__(self):
 
 class RandomRotation(Module):
     """Rotate the image by angle.
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., H, W] shape, where ... means an arbitrary number of leading dimensions.
 
     Args:
@@ -1190,7 +1190,7 @@ def __repr__(self):
 
 class RandomGrayscale(Module):
     """Randomly convert image to grayscale with a probability of p (default 0.1).
-    If the image is flow Tensor, it is expected
+    If the image is oneflow Tensor, it is expected
     to have [..., 3, H, W] shape, where ... means an arbitrary number of leading dimensions
 
     Args:
@@ -1278,14 +1278,14 @@ def __init__(self, kernel_size, sigma=(0.1, 2.0)):
 
     @staticmethod
     def get_params(sigma_min: float, sigma_max: float) -> float:
-        """Choose sigma for random gaussian blurring.
+        """Choose sigma for random Gaussian blurring.
 
         Args:
             sigma_min (float): Minimum standard deviation that can be chosen for blurring kernel.
             sigma_max (float): Maximum standard deviation that can be chosen for blurring kernel.
 
         Returns:
-            float: Standard deviation to be passed to calculate kernel for gaussian blurring.
+            float: Standard deviation to be passed to calculate kernel for Gaussian blurring.
         """
         return flow.empty(1).uniform_(sigma_min, sigma_max).item()
 

flowvision/utils/clip_grad.py

@@ -15,7 +15,7 @@ def dispatch_clip_grad(
 
     Args:
         parameters (Iterable): model parameters to clip
-        value (float): clipping value/factor/norm, mode dependant
+        value (float): clipping value/factor/norm, mode dependent
         mode (str): clipping mode, one of 'norm', 'value', 'agc'
         norm_type (float): p-norm, default 2.0
     """