Fix docs and copyright statements

zhiqwang · Mar 3, 2021 · b6cdb5d · b6cdb5d
1 parent fa192f2
commit b6cdb5d
Show file tree

Hide file tree

Showing 3 changed files with 12 additions and 13 deletions.
diff --git a/yolort/models/backbone_utils.py b/yolort/models/backbone_utils.py
@@ -4,7 +4,6 @@
 
 from . import darknet
 from .path_aggregation_network import PathAggregationNetwork
-from .common import BottleneckCSP, C3
 
 from typing import List, Optional
 
@@ -53,7 +52,7 @@ def darknet_pan_backbone(
     version: str = 'v4.0',
 ):
     """
-    Constructs a specified ResNet backbone with PAN on top. Freezes the specified number of
+    Constructs a specified DarkNet backbone with PAN on top. Freezes the specified number of
     layers in the backbone.
 
     Examples::
@@ -71,12 +70,12 @@ def darknet_pan_backbone(
         >>>    ('2', torch.Size([1, 512, 2, 2]))]
 
     Args:
-        backbone_name (string): resnet architecture. Possible values are 'DarkNet', 'darknet_s_r3_1',
+        backbone_name (string): darknet architecture. Possible values are 'DarkNet', 'darknet_s_r3_1',
            'darknet_m_r3_1', 'darknet_l_r3_1', 'darknet_s_r4_0', 'darknet_m_r4_0', 'darknet_l_r4_0'
         norm_layer (torchvision.ops): it is recommended to use the default value. For details visit:
             (https://github.com/facebookresearch/maskrcnn-benchmark/issues/267)
         pretrained (bool): If True, returns a model with backbone pre-trained on Imagenet
-        trainable_layers (int): number of trainable (not frozen) resnet layers starting from final block.
+        trainable_layers (int): number of trainable (not frozen) darknet layers starting from final block.
             Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable.
         version (str): ultralytics release version: v3.1 or v4.0
     """

diff --git a/yolort/models/yolo.py b/yolort/models/yolo.py
@@ -1,5 +1,4 @@
-# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved
-# Modified by Zhiqiang Wang (me@zhiqwang.com)
+# Copyright (c) 2020, Zhiqiang Wang. All Rights Reserved.
 import warnings
 
 import torch
@@ -306,8 +305,8 @@ def yolov5_darknet_pan_l_r40(pretrained: bool = False, progress: bool = True, nu
 
 def yolov5_darknet_pan_s_tr(pretrained: bool = False, progress: bool = True, num_classes: int = 80,
                             **kwargs: Any) -> YOLO:
-    r"""yolov5 small release 4.0 model from
-    `"ultralytics/yolov5" <https://zenodo.org/badge/latestdoi/264818686>`_.
+    r"""yolov5 small with a transformer block model from
+    `"dingyiwei/yolov5" <https://github.com/ultralytics/yolov5/pull/2333>`_.
     Args:
         pretrained (bool): If True, returns a model pre-trained on ImageNet
         progress (bool): If True, displays a progress bar of the download to stderr

diff --git a/yolort/models/yolotr.py b/yolort/models/yolotr.py
@@ -1,3 +1,4 @@
+# Copyright (c) 2021, Zhiqiang Wang. All Rights Reserved.
 from torch import nn
 
 from .common import Conv, C3
@@ -18,13 +19,13 @@ def darknet_pan_tr_backbone(
     version: str = 'v4.0',
 ):
     """
-    Constructs a specified ResNet backbone with PAN on top. Freezes the specified number of
+    Constructs a specified DarkNet backbone with PAN on top. Freezes the specified number of
     layers in the backbone.
 
     Examples::
 
-        >>> from models.backbone_utils import darknet_pan_backbone
-        >>> backbone = darknet_pan_backbone('darknet3_1', pretrained=True, trainable_layers=3)
+        >>> from models.backbone_utils import darknet_pan_tr_backbone
+        >>> backbone = darknet_pan_tr_backbone('darknet3_1', pretrained=True, trainable_layers=3)
         >>> # get some dummy image
         >>> x = torch.rand(1, 3, 64, 64)
         >>> # compute the output
@@ -36,12 +37,12 @@ def darknet_pan_tr_backbone(
         >>>    ('2', torch.Size([1, 512, 2, 2]))]
 
     Args:
-        backbone_name (string): resnet architecture. Possible values are 'DarkNet', 'darknet_s_r3_1',
+        backbone_name (string): darknet architecture. Possible values are 'DarkNet', 'darknet_s_r3_1',
            'darknet_m_r3_1', 'darknet_l_r3_1', 'darknet_s_r4_0', 'darknet_m_r4_0', 'darknet_l_r4_0'
         norm_layer (torchvision.ops): it is recommended to use the default value. For details visit:
             (https://github.com/facebookresearch/maskrcnn-benchmark/issues/267)
         pretrained (bool): If True, returns a model with backbone pre-trained on Imagenet
-        trainable_layers (int): number of trainable (not frozen) resnet layers starting from final block.
+        trainable_layers (int): number of trainable (not frozen) darknet layers starting from final block.
             Valid values are between 0 and 5, with 5 meaning all backbone layers are trainable.
         version (str): ultralytics release version, currently only supports v3.1 or v4.0
     """