Port torchvision tutorial to gallery and remove previous files

Author: vfdev-5

.jenkins/validate_tutorials_built.py

@@ -22,7 +22,7 @@
     "beginner_source/former_torchies/tensor_tutorial_old",
     "beginner_source/examples_autograd/polynomial_autograd",
     "beginner_source/examples_autograd/polynomial_custom_function",
-    "beginner_source/t5_tutorial", # re-enable after this is fixed: https://github.com/pytorch/text/issues/1756 
+    "beginner_source/t5_tutorial", # re-enable after this is fixed: https://github.com/pytorch/text/issues/1756
     "intermediate_source/parametrizations",
     "intermediate_source/mnist_train_nas",  # used by ax_multiobjective_nas_tutorial.py
     "intermediate_source/fx_conv_bn_fuser",

Makefile

@@ -86,6 +86,9 @@ download:
 	wget -nv -N https://www.manythings.org/anki/deu-eng.zip -P $(DATADIR)
 	unzip -o $(DATADIR)/deu-eng.zip -d beginner_source/data/
 
+	# Download PennFudanPed dataset for intermediate_source/torchvision_tutorial.py
+	wget https://www.cis.upenn.edu/~jshi/ped_html/PennFudanPed.zip -P $(DATADIR)
+	unzip -o $(DATADIR)/PennFudanPed.zip -d intermediate_source/data/
 
 docs:
 	make download
@@ -103,3 +106,5 @@ html-noplot:
 clean-cache:
 	make clean
 	rm -rf advanced beginner intermediate recipes
+	# remove additional python files downloaded for torchvision_tutorial.py
+	rm -rf intermediate_source/engine.py intermediate_source/utils.py intermediate_source/transforms.py intermediate_source/coco_eval.py intermediate_source/coco_utils.py

_static/img/tv_tutorial/tv_image01.png

@@ -10,6 +10,7 @@ RRef
 OOM
 subfolder
 Dialogs
+PennFudan
 performant
 multithreading
 linearities
@@ -30,6 +31,8 @@ breakpoint
 MobileNet
 DeepLabV
 Resampling
+RCNN
+RPN
 APIs
 ATen
 AVX
@@ -134,6 +137,7 @@ LRSchedulers
 Lua
 Luong
 macos
+mAP
 MLP
 MLPs
 MNIST
@@ -165,10 +169,12 @@ OU
 PIL
 PPO
 Plotly
+pre
 Prec
 Profiler
 PyTorch's
 RGB
+RGBA
 RL
 RNN
 RNNs
@@ -331,6 +337,7 @@ jit
 jitter
 jpg
 judgements
+keypoint
 kwargs
 labelled
 learnable
@@ -411,6 +418,7 @@ reinitializes
 relu
 reproducibility
 rescale
+rescaling
 resnet
 restride
 rewinded
@@ -462,10 +470,12 @@ torchscriptable
 torchtext
 torchtext's
 torchvision
+TorchVision
 torchviz
 traceback
 tradeoff
 tradeoffs
+uint
 uncomment
 uncommented
 underflowing

_static/img/tv_tutorial/tv_image02.png

@@ -6,17 +6,10 @@
 
 ######################################################################
 #
-# .. tip::
-#
-#     To get the most of this tutorial, we suggest using this
-#     `Colab Version <https://colab.research.google.com/github/pytorch/tutorials/blob/gh-pages/_downloads/torchvision_finetuning_instance_segmentation.ipynb>`__.
-#     This will allow you to experiment with the information presented below.
-#
-#
 # For this tutorial, we will be finetuning a pre-trained `Mask
-# R-CNN <https://arxiv.org/abs/1703.06870>`__ model on the `Penn-Fudan
+# R-CNN <https://arxiv.org/abs/1703.06870>`_ model on the `Penn-Fudan
 # Database for Pedestrian Detection and
-# Segmentation <https://www.cis.upenn.edu/~jshi/ped_html/>`__. It contains
+# Segmentation <https://www.cis.upenn.edu/~jshi/ped_html/>`_. It contains
 # 170 images with 345 instances of pedestrians, and we will use it to
 # illustrate how to use the new features in torchvision in order to train
 # an object detection and instance segmentation model on a custom dataset.
@@ -65,7 +58,7 @@
 # ``pycocotools`` which can be installed with ``pip install pycocotools``.
 #
 # .. note ::
-#   For Windows, please install ``pycocotools`` from `gautamchitnis <https://github.com/gautamchitnis/cocoapi>`__ with command
+#   For Windows, please install ``pycocotools`` from `gautamchitnis <https://github.com/gautamchitnis/cocoapi>`_ with command
 #
 #   ``pip install git+https://github.com/gautamchitnis/cocoapi.git@cocodataset-master#subdirectory=PythonAPI``
 #
@@ -85,10 +78,16 @@
 # Writing a custom dataset for PennFudan
 # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 #
-# Let’s write a dataset for the PennFudan dataset. After `downloading and
-# extracting the zip
-# file <https://www.cis.upenn.edu/~jshi/ped_html/PennFudanPed.zip>`__, we
-# have the following folder structure:
+# Let’s write a dataset for the PennFudan dataset. First, let's download the dataset and
+# extract the `zip file <https://www.cis.upenn.edu/~jshi/ped_html/PennFudanPed.zip>`_:
+#
+# .. code:: python
+#
+#     wget https://www.cis.upenn.edu/~jshi/ped_html/PennFudanPed.zip -P data
+#     cd data && unzip PennFudanPed.zip
+#
+#
+# We have the following folder structure:
 #
 # ::
 #
@@ -196,8 +195,8 @@ def __len__(self):
 # -------------------
 #
 # In this tutorial, we will be using `Mask
-# R-CNN <https://arxiv.org/abs/1703.06870>`__, which is based on top of
-# `Faster R-CNN <https://arxiv.org/abs/1506.01497>`__. Faster R-CNN is a
+# R-CNN <https://arxiv.org/abs/1703.06870>`_, which is based on top of
+# `Faster R-CNN <https://arxiv.org/abs/1506.01497>`_. Faster R-CNN is a
 # model that predicts both bounding boxes and class scores for potential
 # objects in the image.
 #
@@ -484,7 +483,7 @@ def get_transform(train):
 from torchvision.utils import draw_bounding_boxes, draw_segmentation_masks
 
 
-image = read_image("../_static/img/tv_tutorial/tv_image05.png")
+image = read_image("data/PennFudanPed/PNGImages/FudanPed00046.png")
 eval_transform = get_transform(train=False)
 
 model.eval()
@@ -527,4 +526,4 @@ def get_transform(train):
 # the torchvision repository.
 #
 # You can download a full source file for this tutorial
-# `here <https://pytorch.org/tutorials/_static/tv-training-code.py>`__.
+# `here <https://pytorch.org/tutorials/_static/tv-training-code.py>`_.