Fix typos

lib/cmake/Cuda.cmake

@@ -4,8 +4,8 @@
 # This cmake file is called from Dependencies.cmake. You do not need to
 # manually invoke it.
 
-# Known NVIDIA GPU achitectures Caffe2 can be compiled for.
-# Default is set to cuda 9. If we detect the cuda architectores to be less than
+# Known NVIDIA GPU architectures Caffe2 can be compiled for.
+# Default is set to cuda 9. If we detect the cuda architectures to be less than
 # 9, we will lower it to the corresponding known archs.
 set(Caffe2_known_gpu_archs "30 35 50 52 60 61 70") # for CUDA 9.x
 set(Caffe2_known_gpu_archs8 "20 21(20) 30 35 50 52 60 61") # for CUDA 8.x
@@ -197,7 +197,7 @@ else()
  message(FATAL_ERROR "Cannot find libnvrtc.so. Please file an issue on https://github.com/caffe2/caffe2 with your build output.")
 endif()
 
-# disable some nvcc diagnostic that apears in boost, glog, glags, opencv, etc.
+# disable some nvcc diagnostic that appears in boost, glog, glags, opencv, etc.
 foreach(diag cc_clobber_ignored integer_sign_change useless_using_declaration set_but_not_used)
  list(APPEND CUDA_NVCC_FLAGS -Xcudafe --diag_suppress=${diag})
 endforeach()

lib/core/config.py

@@ -394,7 +394,7 @@
 __C.TEST.KPS_AUG.SCALE_SIZE_DEP = False
 __C.TEST.KPS_AUG.AREA_TH = 180**2
 
-# Eeach aspect ratio is realtive to image width
+# Each aspect ratio is relative to image width
 __C.TEST.KPS_AUG.ASPECT_RATIOS = ()
 
 # Horizontal flip at each aspect ratio
@@ -663,7 +663,7 @@
 # ---------------------------------------------------------------------------- #
 __C.RPN = AttrDict()
 
-# [Infered value; do not set directly in a config]
+# [Inferred value; do not set directly in a config]
 # Indicates that the model contains an RPN subnetwork
 __C.RPN.RPN_ON = False
 
@@ -836,7 +836,7 @@
 # the minibatch is discarded)
 __C.KRCNN.MIN_KEYPOINT_COUNT_FOR_VALID_MINIBATCH = 20
 
-# When infering the keypoint locations from the heatmap, don't scale the heatmap
+# When inferring the keypoint locations from the heatmap, don't scale the heatmap
 # below this minimum size
 __C.KRCNN.INFERENCE_MIN_SIZE = 0
 
@@ -972,7 +972,7 @@
 # If an option is removed from the code and you don't want to break existing
 # yaml configs, you can add the full config key as a string to the set below.
 # ---------------------------------------------------------------------------- #
-_DEPCRECATED_KEYS = set(
+_DEPRECATED_KEYS = set(
  (
  'FINAL_MSG',
  'MODEL.DILATION',
@@ -1118,7 +1118,7 @@ def _merge_a_into_b(a, b, stack=None):
 
 
 def _key_is_deprecated(full_key):
- if full_key in _DEPCRECATED_KEYS:
+ if full_key in _DEPRECATED_KEYS:
  logger.warn(
  'Deprecated config key (ignoring): {}'.format(full_key)
  )

lib/core/test.py

@@ -704,7 +704,7 @@ def im_detect_keypoints_aspect_ratio(
 ):
  """Detects keypoints at the given width-relative aspect ratio."""
 
- # Perform keypoint detectionon the transformed image
+ # Perform keypoint detection on the transformed image
  im_ar = image_utils.aspect_ratio_rel(im, aspect_ratio)
  boxes_ar = box_utils.aspect_ratio(boxes, aspect_ratio)
 

lib/modeling/FPN.py

@@ -96,7 +96,7 @@ def add_fpn_onto_conv_body(
 ):
  """Add the specified conv body to the model and then add FPN levels to it.
  """
- # Note: blobs_conv is in revsersed order: [fpn5, fpn4, fpn3, fpn2]
+ # Note: blobs_conv is in reversed order: [fpn5, fpn4, fpn3, fpn2]
  # similarly for dims_conv: [2048, 1024, 512, 256]
  # similarly for spatial_scales_fpn: [1/32, 1/16, 1/8, 1/4]
 

lib/modeling/detector.py

@@ -102,7 +102,7 @@ def AffineChannel(self, blob_in, blob_out, share_with=None, inplace=False):
  return self.net.AffineChannel([blob_in, scale, bias], blob_out)
 
  def GenerateProposals(self, blobs_in, blobs_out, anchors, spatial_scale):
- """Op for generating RPN porposals.
+ """Op for generating RPN proposals.
 
  blobs_in:
  - 'rpn_cls_probs': 4D tensor of shape (N, A, H, W), where N is the

lib/modeling/name_compat.py

@@ -38,16 +38,16 @@
 
 
 _RENAME = {
- # Removed "ResNet_" from the name because it wasn't relevent
+ # Removed "ResNet_" from the name because it wasn't relevant
  'mask_rcnn_heads.ResNet_mask_rcnn_fcn_head_v1up4convs':
  'mask_rcnn_heads.mask_rcnn_fcn_head_v1up4convs',
- # Removed "ResNet_" from the name because it wasn't relevent
+ # Removed "ResNet_" from the name because it wasn't relevant
  'mask_rcnn_heads.ResNet_mask_rcnn_fcn_head_v1up':
  'mask_rcnn_heads.mask_rcnn_fcn_head_v1up',
- # Removed "ResNet_" from the name because it wasn't relevent
+ # Removed "ResNet_" from the name because it wasn't relevant
  'mask_rcnn_heads.ResNet_mask_rcnn_fcn_head_v0upshare':
  'mask_rcnn_heads.mask_rcnn_fcn_head_v0upshare',
- # Removed "ResNet_" from the name because it wasn't relevent
+ # Removed "ResNet_" from the name because it wasn't relevant
  'mask_rcnn_heads.ResNet_mask_rcnn_fcn_head_v0up':
  'mask_rcnn_heads.mask_rcnn_fcn_head_v0up',
  # Removed head_builder module in favor of the more specific fast_rcnn name

lib/ops/generate_proposals.py

@@ -55,7 +55,7 @@ def forward(self, inputs, outputs):
 
  # predicted probability of fg object for each RPN anchor
  scores = inputs[0].data
- # predicted achors transformations
+ # predicted anchors transformations
  bbox_deltas = inputs[1].data
  # input image (height, width, scale), in which scale is the scale factor
  # applied to the original dataset image to get the network input image
@@ -71,7 +71,7 @@ def forward(self, inputs, outputs):
  shifts = np.vstack((shift_x.ravel(), shift_y.ravel(),
  shift_x.ravel(), shift_y.ravel())).transpose()
 
- # Broacast anchors over shifts to enumerate all anchors at all positions
+ # Broadcast anchors over shifts to enumerate all anchors at all positions
  # in the (H, W) grid:
  # - add A anchors of shape (1, A, 4) to
  # - K shifts of shape (K, 1, 4) to get

lib/roi_data/data_utils.py

@@ -13,7 +13,7 @@
 # limitations under the License.
 ##############################################################################
 
-"""Common utility functions for RPN and RetinaNet minibtach blobs preparation.
+"""Common utility functions for RPN and RetinaNet minibatch blobs preparation.
 """
 
 from __future__ import absolute_import
@@ -76,7 +76,7 @@ def get_field_of_anchors(
  shift_y = shift_y.ravel()
  shifts = np.vstack((shift_x, shift_y, shift_x, shift_y)).transpose()
 
- # Broacast anchors over shifts to enumerate all anchors at all positions
+ # Broadcast anchors over shifts to enumerate all anchors at all positions
  # in the (H, W) grid:
  # - add A cell anchors of shape (1, A, 4) to
  # - K shifts of shape (K, 1, 4) to get

lib/roi_data/fast_rcnn.py

@@ -15,7 +15,7 @@
 
 """Construct minibatches for Fast R-CNN training. Handles the minibatch blobs
 that are specific to Fast R-CNN. Other blobs that are generic to RPN, etc.
-are handled by their respecitive roi_data modules.
+are handled by their respective roi_data modules.
 """
 
 from __future__ import absolute_import
@@ -52,7 +52,7 @@ def get_fast_rcnn_blob_names(is_training=True):
  # targets per class
  blob_names += ['bbox_targets']
  # bbox_inside_weights blob: At most 4 targets per roi are active
- # this binary vector sepcifies the subset of active targets
+ # this binary vector specifies the subset of active targets
  blob_names += ['bbox_inside_weights']
  blob_names += ['bbox_outside_weights']
  if is_training and cfg.MODEL.MASK_ON:

lib/roi_data/keypoint_rcnn.py

@@ -16,7 +16,7 @@
 """Construct minibatches for Mask R-CNN training when keypoints are enabled.
 Handles the minibatch blobs that are specific to training Mask R-CNN for
 keypoint detection. Other blobs that are generic to RPN or Fast/er R-CNN are
-handled by their respecitive roi_data modules.
+handled by their respective roi_data modules.
 """
 
 from __future__ import absolute_import

lib/roi_data/mask_rcnn.py

@@ -15,7 +15,7 @@
 
 """Construct minibatches for Mask R-CNN training. Handles the minibatch blobs
 that are specific to Mask R-CNN. Other blobs that are generic to RPN or
-Fast/er R-CNN are handled by their respecitive roi_data modules.
+Fast/er R-CNN are handled by their respective roi_data modules.
 """
 
 from __future__ import absolute_import

lib/roi_data/retinanet.py

@@ -130,7 +130,7 @@ def add_retinanet_blobs(blobs, im_scales, roidb, image_width, image_height):
  loc_stride *= (cfg.MODEL.NUM_CLASSES - 1)
  anchor_ind = foa.octave * num_aspect_ratios + foa.aspect
  # v[:, 1] is the class label [range 0-80] if we do
- # class-specfic bbox otherwise it is 0. In case of class
+ # class-specific bbox otherwise it is 0. In case of class
  # specific, based on the label, the location of current
  # anchor is class_label * 4 and then we take into account
  # the anchor_ind if the anchors
@@ -153,7 +153,7 @@ def add_retinanet_blobs(blobs, im_scales, roidb, image_width, image_height):
  # and length of this list is N x A where
  # N = num_images, A = num_anchors for example, N = 2, A = 9
  # Each element of the list has the shape 1 x 1 x H x W where H, W are
- # spatial dimension of curret fpn lvl. Let a{i} denote the element
+ # spatial dimension of current fpn lvl. Let a{i} denote the element
  # corresponding to anchor i [9 anchors total] in the list.
  # The elements in the list are in order [[a0, ..., a9], [a0, ..., a9]]
  # however the network will make predictions like 2 x (9 * 80) x H x W

lib/utils/keypoints.py

@@ -28,7 +28,7 @@
 
 
 def get_keypoints():
- """Get the COCO keypoints and their left/right flip coorespondence map."""
+ """Get the COCO keypoints and their left/right flip correspondence map."""
  # Keypoints are not available in the COCO json for the test split, so we
  # provide them here.
  keypoints = [

lib/utils/segms.py

@@ -199,7 +199,7 @@ def rle_mask_nms(masks, dets, thresh, mode='IOU'):
  """Performs greedy non-maximum suppression based on an overlap measurement
  between masks. The type of measurement is determined by `mode` and can be
  either 'IOU' (standard intersection over union) or 'IOMA' (intersection over
- mininum area).
+ minimum area).
  """
  if len(masks) == 0:
  return []

tools/pickle_caffe_blobs.py

@@ -16,7 +16,7 @@
 ##############################################################################
 
 """Script for converting Caffe (<= 1.0) models into the the simple state dict
-format used by Detectron. For example, this script can convert the orignal
+format used by Detectron. For example, this script can convert the original
 ResNet models released by MSRA.
 """
 

tools/train_net.py

@@ -189,7 +189,7 @@ def main():
  logger.info(pprint.pformat(cfg))
  # Note that while we set the numpy random seed network training will not be
  # deterministic in general. There are sources of non-determinism that cannot
- # be removed with a reasonble execution-speed tradeoff (such as certain
+ # be removed with a reasonable execution-speed tradeoff (such as certain
  # non-deterministic cudnn functions).
  np.random.seed(cfg.RNG_SEED)
  # Execute the training run