Merge branch 'pytorch:main' into main

Author: ZhaoqiongZ

.ci/docker/Dockerfile

@@ -15,15 +15,11 @@ RUN bash ./install_user.sh && rm install_user.sh
 COPY ./common/install_docs_reqs.sh install_docs_reqs.sh
 RUN bash ./install_docs_reqs.sh && rm install_docs_reqs.sh
 
-# Install conda and other packages
-ENV ANACONDA_PYTHON_VERSION=3.10
-ENV CONDA_CMAKE yes
-ENV DOCS yes
-ENV PATH /opt/conda/envs/py_$ANACONDA_PYTHON_VERSION/bin:/opt/conda/bin:$PATH
-COPY ./requirements.txt /opt/conda/
-COPY ./common/install_conda.sh install_conda.sh
-COPY ./common/common_utils.sh common_utils.sh
-RUN bash ./install_conda.sh && rm install_conda.sh common_utils.sh /opt/conda/requirements.txt
+COPY ./common/install_pip_requirements.sh install_pip_requirements.sh
+COPY ./requirements.txt requirements.txt
+RUN bash ./install_pip_requirements.sh && rm install_pip_requirements.sh
+
+RUN ln -s /usr/bin/python3 /usr/bin/python
 
 USER ci-user
 CMD ["bash"]

.ci/docker/build.sh

@@ -10,7 +10,7 @@ set -exu
 IMAGE_NAME="$1"
 shift
 
-export UBUNTU_VERSION="20.04"
+export UBUNTU_VERSION="22.04"
 export CUDA_VERSION="12.4.1"
 
 export BASE_IMAGE="nvidia/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION}"

.ci/docker/common/common_utils.sh

@@ -10,7 +10,7 @@ install_ubuntu() {
   apt-get install -y --no-install-recommends \
     build-essential \
     ca-certificates \
-    cmake=3.16* \
+    cmake=3.22* \
     curl \
     git \
     wget \
@@ -27,7 +27,9 @@ install_ubuntu() {
     libglfw3-dev \
     sox \
     libsox-dev \
-    libsox-fmt-all
+    libsox-fmt-all \
+    python3-pip \
+    python3-dev
 
   # Cleanup package manager
   apt-get autoclean && apt-get clean

.ci/docker/common/install_base.sh

@@ -0,0 +1,7 @@
+#!/bin/bash
+
+set -ex
+
+# Install pip packages
+pip install --upgrade pip
+pip install -r ./requirements.txt

.ci/docker/common/install_conda.sh

@@ -14,7 +14,7 @@ tqdm==4.66.1
 numpy==1.24.4
 matplotlib
 librosa
-torch==2.4
+torch==2.6
 torchvision
 torchdata
 networkx
@@ -28,8 +28,8 @@ tensorboard
 jinja2==3.1.3
 pytorch-lightning
 torchx
-torchrl==0.5.0
-tensordict==0.5.0
+torchrl==0.7.2
+tensordict==0.7.2
 ax-platform>=0.4.0
 nbformat>=5.9.2
 datasets
@@ -69,5 +69,5 @@ pycocotools
 semilearn==0.3.2
 torchao==0.5.0
 segment_anything==1.0
-torchrec==0.8.0
-fbgemm-gpu==0.8.0
+torchrec==1.0.0; platform_system == "Linux"
+fbgemm-gpu==1.1.0; platform_system == "Linux"

.ci/docker/common/install_pip_requirements.sh

@@ -0,0 +1,44 @@
+#Runs once a month and checks links in the repo to ensure they are valid 
+#If action fails, it creates an issue with the failing links and an "incorrect link" label
+#If link is valid but failing, it can be added to the .lycheeignore file
+#Action can also be run manually as needed.
+
+
+name: Monthly Link Check
+on:
+  schedule:
+    - cron: '0 0 1 * *'  # Runs at midnight on the first day of every month
+  workflow_dispatch:  # Allows manual triggering of the workflow
+jobs:
+  linkChecker:
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+      - name: Check Links
+        id: lychee
+        uses: lycheeverse/lychee-action@v2
+        with:
+          args: --accept=200,403,429 --base . --verbose --no-progress './**/*.md' './**/*.html' './**/*.rst'
+          token: ${{ secrets.CUSTOM_TOKEN }}
+          fail: true
+
+      - name: Create Issue From File
+        if: failure() && (github.event_name == 'schedule' || github.event_name == 'workflow_dispatch')
+        uses: peter-evans/create-issue-from-file@v5
+        with:
+          title: Broken links detected in docs 🔗
+          content-filepath: ./lychee/out.md
+          labels: 'incorrect link'
+          #token: ${{ secrets.CUSTOM_TOKEN }}
+          
+    
+      - name: Suggestions
+        if: failure()
+        run: |
+          echo -e "\nPlease review the links reported in the Check links step above."
+          echo -e "If a link is valid but fails due to a CAPTCHA challenge, IP blocking, login requirements, etc., consider adding such links to .lycheeignore file to bypass future checks.\n"
+          exit 1

.ci/docker/requirements.txt

@@ -57,11 +57,11 @@ jobs:
             const STALE_CLOSE_THRESHOLD_MS = 1000 * 60 * 60 * 24 * 30;
 
             const STALE_MESSAGE =
-              "Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as `Stale`. <br>" +
-              "Feel free to remove the `Stale` label if you feel this was a mistake. <br>" +
-              "If you are unable to remove the `Stale` label please contact a maintainer in order to do so. <br>" +
+              "Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as `stale`. <br>" +
+              "Feel free to remove the `stale` label if you feel this was a mistake. <br>" +
+              "If you are unable to remove the `stale` label please contact a maintainer in order to do so. <br>" +
               "If you want the bot to never mark this PR stale again, add the `no-stale` label.<br>" +
-              "`Stale` pull requests will automatically be closed after 30 days of inactivity.<br>";
+              "`stale` pull requests will automatically be closed after 30 days of inactivity.<br>";
 
             let numAPIRequests = 0;
             let numProcessed = 0;
@@ -79,7 +79,7 @@ jobs:
 
               // Check if the PR is stale, according to our configured thresholds.
               let staleThresholdMillis;
-              if (labels.includes("Stale")) {
+              if (labels.includes("stale")) {
                 core.info(`[${pull.number}] PR is labeled stale, checking whether we should close it.`);
                 staleThresholdMillis = STALE_CLOSE_THRESHOLD_MS;
               } else {
@@ -97,7 +97,7 @@ jobs:
 
               // At this point, we know we should do something.
               // For PRs already labeled stale, close them.
-              if (labels.includes("Stale")) {
+              if (labels.includes("stale")) {
                 core.info(`[${pull.number}] Closing PR.`);
                 numAPIRequests += 1;
                 await github.rest.issues.update({
@@ -123,7 +123,7 @@ jobs:
                   owner: "pytorch",
                   repo: "tutorials",
                   issue_number: pull.number,
-                  labels: ["Stale"],
+                  labels: ["stale"],
                 });
               }
             }

.github/workflows/MonthlyLinkCheck.yml

@@ -44,6 +44,8 @@ jobs:
 
       - name: Checkout Tutorials
         uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
 
       - name: Setup Linux
         uses: pytorch/pytorch/.github/actions/setup-linux@main
@@ -115,6 +117,8 @@ jobs:
 
       - name: Checkout Tutorials
         uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
 
       - name: Setup Linux
         uses: pytorch/pytorch/.github/actions/setup-linux@main