[BugFix] Fix Examples (pytorch#687)

ai4co · Nov 21, 2022 · a3bbba0 · a3bbba0
1 parent 40c04ef
commit a3bbba0
Show file tree

Hide file tree

Showing 22 changed files with 774 additions and 51 deletions.
diff --git a/.circleci/config.yml b/.circleci/config.yml
@@ -347,6 +347,57 @@ jobs:
       - store_test_results:
           path: test-results
 
+
+  unittest_linux_examples_gpu:
+    <<: *binary_common
+    machine:
+      image: ubuntu-2004-cuda-11.4:202110-01
+    resource_class: gpu.nvidia.medium
+    environment:
+      image_name: "pytorch/manylinux-cuda113"
+      TAR_OPTIONS: --no-same-owner
+      PYTHON_VERSION: << parameters.python_version >>
+      CU_VERSION: << parameters.cu_version >>
+
+    steps:
+      - checkout
+      - designate_upload_channel
+      - run:
+          name: Generate cache key
+          # This will refresh cache on Sundays, nightly build should generate new cache.
+          command: echo "$(date +"%Y-%U")" > .circleci-weekly
+      - restore_cache:
+
+          keys:
+            - env-v3-linux-{{ arch }}-py<< parameters.python_version >>-{{ checksum ".circleci/unittest/linux_examples/scripts/environment.yml" }}-{{ checksum ".circleci-weekly" }}
+
+      - run:
+          name: Setup
+          command: docker run -e PYTHON_VERSION -t --gpus all -v $PWD:$PWD -w $PWD "${image_name}" .circleci/unittest/linux_examples/scripts/setup_env.sh
+      - save_cache:
+
+          key: env-v3-linux-{{ arch }}-py<< parameters.python_version >>-{{ checksum ".circleci/unittest/linux_examples/scripts/environment.yml" }}-{{ checksum ".circleci-weekly" }}
+
+          paths:
+            - conda
+            - env
+      - run:
+          name: Install torchrl
+          command: docker run -t --gpus all -v $PWD:$PWD -w $PWD -e UPLOAD_CHANNEL -e CU_VERSION "${image_name}" .circleci/unittest/linux_examples/scripts/install.sh
+      - run:
+          name: Run tests
+          command: bash .circleci/unittest/linux_examples/scripts/run_test.sh
+#          command: docker run --env-file ./env.list -t --gpus all -v $PWD:$PWD -w $PWD "${image_name}" .circleci/unittest/linux/scripts/run_test.sh
+      - run:
+          name: Codecov upload
+          command: |
+            bash <(curl -s https://codecov.io/bash) -Z -F linux_examples-gpu
+      - run:
+          name: Post Process
+          command: docker run -t --gpus all -v $PWD:$PWD -w $PWD "${image_name}" .circleci/unittest/linux_examples/scripts/post_process.sh
+      - store_test_results:
+          path: test-results
+
   unittest_linux_habitat_gpu:
     <<: *binary_common
     machine:
@@ -937,3 +988,8 @@ workflows:
           cu_version: cu113
           name: unittest_linux_olddeps_gpu_py3.9
           python_version: '3.9'
+
+      - unittest_linux_examples_gpu:
+          cu_version: cu113
+          name: unittest_linux_examples_gpu_py3.9
+          python_version: '3.9'
diff --git a/.circleci/unittest/linux_examples/scripts/environment.yml b/.circleci/unittest/linux_examples/scripts/environment.yml
@@ -0,0 +1,30 @@
+channels:
+  - pytorch
+  - defaults
+dependencies:
+  - pip
+  - protobuf
+  - pip:
+    - hypothesis
+    - future
+    - cloudpickle
+    - gym
+    - pygame
+    - gym[accept-rom-license]
+    - gym[atari]
+    - moviepy
+    - tqdm
+    - pytest
+    - pytest-cov
+    - pytest-mock
+    - pytest-instafail
+    - expecttest
+    - pyyaml
+    - scipy
+    - hydra-core
+    - tensorboard
+    - wandb
+    - dm_control
+    - mlflow
+    - av
+    - coverage
diff --git a/.circleci/unittest/linux_examples/scripts/install.sh b/.circleci/unittest/linux_examples/scripts/install.sh
@@ -0,0 +1,46 @@
+#!/usr/bin/env bash
+
+unset PYTORCH_VERSION
+# For unittest, nightly PyTorch is used as the following section,
+# so no need to set PYTORCH_VERSION.
+# In fact, keeping PYTORCH_VERSION forces us to hardcode PyTorch version in config.
+
+set -e
+
+eval "$(./conda/bin/conda shell.bash hook)"
+conda activate ./env
+
+if [ "${CU_VERSION:-}" == cpu ] ; then
+    version="cpu"
+    echo "Using cpu build"
+else
+    if [[ ${#CU_VERSION} -eq 4 ]]; then
+        CUDA_VERSION="${CU_VERSION:2:1}.${CU_VERSION:3:1}"
+    elif [[ ${#CU_VERSION} -eq 5 ]]; then
+        CUDA_VERSION="${CU_VERSION:2:2}.${CU_VERSION:4:1}"
+    fi
+    echo "Using CUDA $CUDA_VERSION as determined by CU_VERSION ($CU_VERSION)"
+    version="$(python -c "print('.'.join(\"${CUDA_VERSION}\".split('.')[:2]))")"
+fi
+
+# submodules
+git submodule sync && git submodule update --init --recursive
+
+printf "Installing PyTorch with %s\n" "${CU_VERSION}"
+if [ "${CU_VERSION:-}" == cpu ] ; then
+    pip3 install --pre torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/nightly/cpu
+else
+    pip3 install --pre torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/nightly/cu113
+fi
+
+# smoke test
+python -c "import functorch"
+
+# install snapshot
+pip install git+https://github.com/pytorch/torchsnapshot
+
+# install tensordict
+pip install git+https://github.com/pytorch-labs/tensordict
+
+printf "* Installing torchrl\n"
+python setup.py develop
diff --git a/.circleci/unittest/linux_examples/scripts/post_process.sh b/.circleci/unittest/linux_examples/scripts/post_process.sh
@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+
+set -e
+
+eval "$(./conda/bin/conda shell.bash hook)"
+conda activate ./env