CI: check dead links in docs

.actions/helpers.py

@@ -79,7 +79,7 @@
 #
 # ### Great thanks from the entire Pytorch Lightning Team for your interest !
 #
-# ![Pytorch Lightning](https://github.com/PyTorchLightning/pytorch-lightning/blob/master/docs/source/_static/images/logo.png){height="60px" width="240px"}
+# [![Pytorch Lightning](https://raw.githubusercontent.com/PyTorchLightning/pytorch-lightning/master/docs/source/_static/images/logo.png){height="60px" width="240px"}](https://pytorchlightning.ai)
 
 """
 TEMPLATE_CARD_ITEM = """

.azure-pipelines/ipynb-publish.yml

@@ -17,7 +17,7 @@ jobs:
  # - For 60 minutes on Microsoft-hosted agents with a private project or private repository
  timeoutInMinutes: 0
 
- pool: gridai-spot-pool
+ pool: azure-gpus-persist
  # this need to have installed docker in the base image...
  container:
  # base ML image: mcr.microsoft.com/azureml/openmpi3.1.2-cuda10.2-cudnn8-ubuntu18.04

.azure-pipelines/ipynb-tests.yml

@@ -13,7 +13,7 @@ jobs:
  # how much time to give 'run always even if cancelled tasks' before stopping them
  cancelTimeoutInMinutes: 2
 
- pool: gridai-spot-pool
+ pool: azure-gpus-spot
  # this need to have installed docker in the base image...
  container:
  # base ML image: mcr.microsoft.com/azureml/openmpi3.1.2-cuda10.2-cudnn8-ubuntu18.04

.github/workflows/ci_docs.yml

@@ -1,4 +1,4 @@
-name: test Docs
+name: validate Docs
 
 on: # Trigger the workflow on push or pull request
  pull_request: {}
@@ -71,7 +71,7 @@ jobs:
  working-directory: ./docs
  run: |
  # First run the same pipeline as Read-The-Docs
- make html --debug --jobs $(nproc) SPHINXOPTS="-W --keep-going"
+ make html --debug --jobs $(nproc) SPHINXOPTS="-W --keep-going" -b linkcheck
 
  - name: Upload built docs
  uses: actions/upload-artifact@v2

course_UvA-DL/01-introduction-to-pytorch/.meta.yml

@@ -1,7 +1,7 @@
 title: "Tutorial 1: Introduction to PyTorch"
 author: Phillip Lippe
 created: 2021-08-27
-updated: 2021-08-27
+updated: 2021-11-29
 license: CC BY-SA
 description: |
  This tutorial will give a short introduction to PyTorch basics, and get you setup for writing your own neural networks.

course_UvA-DL/01-introduction-to-pytorch/Introduction_to_PyTorch.py

@@ -4,7 +4,7 @@
 # The following notebook is meant to give a short introduction to PyTorch basics, and get you setup for writing your own neural networks.
 # PyTorch is an open source machine learning framework that allows you to write your own neural networks and optimize them efficiently.
 # However, PyTorch is not the only framework of its kind.
-# Alternatives to PyTorch include [TensorFlow](https://www.tensorflow.org/), [JAX](https://github.com/google/jax#quickstart-colab-in-the-cloud) and [Caffe](http://caffe.berkeleyvision.org/).
+# Alternatives to PyTorch include [TensorFlow](https://www.tensorflow.org/), [JAX](https://github.com/google/jax) and [Caffe](http://caffe.berkeleyvision.org/).
 # We choose to teach PyTorch at the University of Amsterdam because it is well established, has a huge developer community (originally developed by Facebook), is very flexible and especially used in research.
 # Many current papers publish their code in PyTorch, and thus it is good to be familiar with PyTorch as well.
 # Meanwhile, TensorFlow (developed by Google) is usually known for being a production-grade deep learning library.

course_UvA-DL/03-initialization-and-optimization/.meta.yml

@@ -1,7 +1,7 @@
 title: "Tutorial 3: Initialization and Optimization"
 author: Phillip Lippe
 created: 2021-08-27
-updated: 2021-08-27
+updated: 2021-11-29
 license: CC BY-SA
 tags:
  - Image

course_UvA-DL/03-initialization-and-optimization/Initialization_and_Optimization.py

@@ -323,7 +323,7 @@ def visualize_activations(model, color="C0", print_variance=False):
 # %% [markdown]
 # ## Initialization
 #
-# Before starting our discussion about initialization, it should be noted that there exist many very good blog posts about the topic of neural network initialization (for example [deeplearning.ai](https://www.deeplearning.ai/ai-notes/initialization/), or a more [math-focused blog post](https://pouannes.github.io/blog/initialization/#mjx-eqn-eqfwd_K)).
+# Before starting our discussion about initialization, it should be noted that there exist many very good blog posts about the topic of neural network initialization (for example [deeplearning.ai](https://www.deeplearning.ai/ai-notes/initialization/), or a more [math-focused blog post](https://pouannes.github.io/blog/initialization)).
 # In case something remains unclear after this tutorial, we recommend skimming through these blog posts as well.
 #
 # When initializing a neural network, there are a few properties we would like to have.
@@ -457,7 +457,7 @@ def equal_var_init(model):
 # Besides the variance of the activations, another variance we would like to stabilize is the one of the gradients.
 # This ensures a stable optimization for deep networks.
 # It turns out that we can do the same calculation as above starting from $\Delta x=W\Delta y$, and come to the conclusion that we should initialize our layers with $1/d_y$ where $d_y$ is the number of output neurons.
-# You can do the calculation as a practice, or check a thorough explanation in [this blog post](https://pouannes.github.io/blog/initialization/#mjx-eqn-eqfwd_K).
+# You can do the calculation as a practice, or check a thorough explanation in [this blog post](https://pouannes.github.io/blog/initialization).
 # As a compromise between both constraints, [Glorot and Bengio (2010)](http://proceedings.mlr.press/v9/glorot10a/glorot10a.pdf?hc_location=ufi) proposed to use the harmonic mean of both values.
 # This leads us to the well-known Xavier initialization:
 #

course_UvA-DL/04-inception-resnet-densenet/.meta.yaml

@@ -1,7 +1,7 @@
 title: "Tutorial 4: Inception, ResNet and DenseNet"
 author: Phillip Lippe
 created: 2021-08-27
-updated: 2021-08-27
+updated: 2021-11-29
 license: CC BY-SA
 tags:
  - Image

course_UvA-DL/04-inception-resnet-densenet/Inception_ResNet_DenseNet.py

@@ -208,7 +208,7 @@
 # 5. Test loop (`test_step`) which is the same as validation, only on a test set.
 #
 # Therefore, we don't abstract the PyTorch code, but rather organize it and define some default operations that are commonly used.
-# If you need to change something else in your training/validation/test loop, there are many possible functions you can overwrite (see the [docs](https://pytorch-lightning.readthedocs.io/en/stable/lightning_module.html) for details).
+# If you need to change something else in your training/validation/test loop, there are many possible functions you can overwrite (see the [docs](https://pytorch-lightning.readthedocs.io/en/stable/common/lightning_module.html) for details).
 #
 # Now we can look at an example of how a Lightning Module for training a CNN looks like:
 
@@ -322,7 +322,7 @@ def create_model(model_name, model_hparams):
 # Besides the Lightning module, the second most important module in PyTorch Lightning is the `Trainer`.
 # The trainer is responsible to execute the training steps defined in the Lightning module and completes the framework.
 # Similar to the Lightning module, you can override any key part that you don't want to be automated, but the default settings are often the best practice to do.
-# For a full overview, see the [documentation](https://pytorch-lightning.readthedocs.io/en/stable/trainer.html).
+# For a full overview, see the [documentation](https://pytorch-lightning.readthedocs.io/en/stable/common/trainer.html).
 # The most important functions we use below are:
 #
 # * `trainer.fit`: Takes as input a lightning module, a training dataset, and an (optional) validation dataset.
@@ -764,7 +764,7 @@ def forward(self, x):
 #
 # The three groups operate on the resolutions $32\times32$, $16\times16$ and $8\times8$ respectively.
 # The blocks in orange denote ResNet blocks with downsampling.
-# The same notation is used by many other implementations such as in the [torchvision library](https://pytorch.org/docs/stable/_modules/torchvision/models/resnet.html#resnet18) from PyTorch.
+# The same notation is used by many other implementations such as in the [torchvision library](https://pytorch.org/vision/0.11/models.html#torchvision.models.resnet18) from PyTorch.
 # Thus, our code looks as follows:
 
 

course_UvA-DL/05-transformers-and-MH-attention/.meta.yml

@@ -1,7 +1,7 @@
 title: "Tutorial 5: Transformers and Multi-Head Attention"
 author: Phillip Lippe
 created: 2021-06-30
-updated: 2021-06-30
+updated: 2021-11-29
 license: CC BY-SA
 build: 0
 tags:

course_UvA-DL/05-transformers-and-MH-attention/Transformers_MHAttention.py

@@ -669,10 +669,9 @@ def forward(self, x):
 # Improved optimizers like [RAdam](https://arxiv.org/abs/1908.03265) have been shown to overcome this issue,
 # not requiring warm-up for training Transformers.
 # Secondly, the iteratively applied Layer Normalization across layers can lead to very high gradients during
-# the first iterations, which can be solved by using
-# [Pre-Layer Normalization](https://proceedings.icml.cc/static/paper_files/icml/2020/328-Paper.pdf)
+# the first iterations, which can be solved by using Pre-Layer Normalization
 # (similar to Pre-Activation ResNet), or replacing Layer Normalization by other techniques
-# ([Adaptive Normalization](https://proceedings.icml.cc/static/paper_files/icml/2020/328-Paper.pdf),
+# (Adaptive Normalization,
 # [Power Normalization](https://arxiv.org/abs/2003.07845)).
 #
 # Nevertheless, many applications and papers still use the original Transformer architecture with Adam,

course_UvA-DL/06-graph-neural-networks/GNN_overview.py

@@ -805,7 +805,7 @@ def print_results(result_dict):
 # Torch geometric uses a different, more efficient approach: we can view the $N$ graphs in a batch as a single large graph with concatenated node and edge list.
 # As there is no edge between the $N$ graphs, running GNN layers on the large graph gives us the same output as running the GNN on each graph separately.
 # Visually, this batching strategy is visualized below (figure credit - PyTorch Geometric team,
-# [tutorial here](https://colab.research.google.com/drive/1I8a0DfQ3fI7Njc62__mVXUlcAleUclnb?usp=sharing#scrollTo=2owRWKcuoALo)).
+# [tutorial here](https://colab.research.google.com/drive/1I8a0DfQ3fI7Njc62__mVXUlcAleUclnb)).
 #
 # <center width="100%"><img src="torch_geometric_stacking_graphs.png" width="600px"></center>
 #

flash_tutorials/electricity_forecasting/.meta.yml

@@ -3,7 +3,7 @@ author: Ethan Harris (ethan@pytorchlightning.ai)
 created: 2021-11-23
 updated: 2021-11-23
 license: CC BY-SA
-build: 2
+build: 3
 tags:
  - Tabular
  - Forecasting

lightning_examples/augmentation_kornia/.meta.yml

@@ -3,7 +3,7 @@ author: PL/Kornia team
 created: 2021-06-11
 updated: 2021-06-16
 license: CC BY-SA
-build: 2
+build: 3
 tags:
  - Image
 description: |

lightning_examples/basic-gan/.meta.yaml

@@ -3,7 +3,7 @@ author: PL team
 created: 2020-12-21
 updated: 2021-06-16
 license: CC BY-SA
-build: 3
+build: 4
 tags:
  - Image
 description: |

lightning_examples/cifar10-baseline/.meta.yml

@@ -3,7 +3,7 @@ author: PL team
 created: 2020-12-21
 updated: 2021-06-16
 license: CC BY-SA
-build: 1
+build: 2
 tags:
  - Image
 description: >

lightning_examples/datamodules/.meta.yml

@@ -3,7 +3,7 @@ author: PL team
 created: 2020-12-21
 updated: 2021-06-07
 license: CC BY-SA
-build: 1
+build: 2
 description: This notebook will walk you through how to start using Datamodules. With
  the release of `pytorch-lightning` version 0.9.0, we have included a new class called
  `LightningDataModule` to help you decouple data related hooks from your `LightningModule`.

lightning_examples/datamodules/datamodules.py

@@ -9,9 +9,9 @@
 import torch
 import torch.nn.functional as F
 from pytorch_lightning import LightningDataModule, LightningModule, Trainer
-from pytorch_lightning.metrics.functional import accuracy
 from torch import nn
 from torch.utils.data import DataLoader, random_split
+from torchmetrics.functional import accuracy
 from torchvision import transforms
 
 # Note - you must have torchvision installed for this example

lightning_examples/mnist-hello-world/.meta.yml

@@ -3,7 +3,7 @@ author: PL team
 created: 2020-12-21
 updated: 2021-06-16
 license: CC BY-SA
-build: 1
+build: 2
 tags:
  - Image
 description: In this notebook, we'll go over the basics of lightning by preparing

lightning_examples/mnist-hello-world/hello-world.py

@@ -79,17 +79,17 @@ def configure_optimizers(self):
 #
 # ### Note what the following built-in functions are doing:
 #
-# 1. [prepare_data()](https://pytorch-lightning.readthedocs.io/en/latest/api/pytorch_lightning.core.lightning.html#pytorch_lightning.core.lightning.LightningModule.prepare_data) 💾
+# 1. [prepare_data()](https://pytorch-lightning.readthedocs.io/en/stable/common/lightning_module.html#prepare-data) 💾
 # - This is where we can download the dataset. We point to our desired dataset and ask torchvision's `MNIST` dataset class to download if the dataset isn't found there.
 # - **Note we do not make any state assignments in this function** (i.e. `self.something = ...`)
 #
-# 2. [setup(stage)](https://pytorch-lightning.readthedocs.io/en/latest/common/lightning-module.html#setup) ⚙️
+# 2. [setup(stage)](https://pytorch-lightning.readthedocs.io/en/stable/common/lightning_module.html#setup) ⚙️
 # - Loads in data from file and prepares PyTorch tensor datasets for each split (train, val, test).
 # - Setup expects a 'stage' arg which is used to separate logic for 'fit' and 'test'.
 # - If you don't mind loading all your datasets at once, you can set up a condition to allow for both 'fit' related setup and 'test' related setup to run whenever `None` is passed to `stage` (or ignore it altogether and exclude any conditionals).
 # - **Note this runs across all GPUs and it *is* safe to make state assignments here**
 #
-# 3. [x_dataloader()](https://pytorch-lightning.readthedocs.io/en/latest/common/lightning-module.html#data-hooks) ♻️
+# 3. [x_dataloader()](https://pytorch-lightning.readthedocs.io/en/stable/api/pytorch_lightning.core.hooks.html) ♻️
 # - `train_dataloader()`, `val_dataloader()`, and `test_dataloader()` all return PyTorch `DataLoader` instances that are created by wrapping their respective datasets that we prepared in `setup()`
 
 

lightning_examples/mnist-tpu-training/.meta.yml

@@ -3,7 +3,7 @@ author: PL team
 created: 2020-12-21
 updated: 2021-06-25
 license: CC BY-SA
-build: 0
+build: 1
 tags:
  - Image
 description: In this notebook, we'll train a model on TPUs. Updating one Trainer flag is all you need for that.

lightning_examples/reinforce-learning-DQN/.meta.yml

@@ -3,7 +3,7 @@ author: PL team
 created: 2021-01-31
 updated: 2021-06-17
 license: CC BY-SA
-build: 1
+build: 2
 tags:
  - RL
 description: |

lightning_examples/text-transformers/text-transformers.py

@@ -274,7 +274,7 @@ def configure_optimizers(self):
 )
 
 trainer = Trainer(max_epochs=1, gpus=AVAIL_GPUS)
-trainer.fit(model, dm)
+trainer.fit(model, datamodule=dm)
 
 # %% [markdown]
 # ### MRPC
@@ -298,7 +298,7 @@ def configure_optimizers(self):
 )
 
 trainer = Trainer(max_epochs=3, gpus=AVAIL_GPUS)
-trainer.fit(model, dm)
+trainer.fit(model, datamodule=dm)
 
 # %% [markdown]
 # ### MNLI

sample-template/.meta.yml

@@ -3,7 +3,7 @@ author: PL team
 created: 2021-06-15
 updated: 2021-06-17
 license: CC
-build: 4
+build: 5
 description: |
  This is a template to show how to contribute a tutorial.
 requirements: