Merge recent changes in master into release branch (#799)

Co-authored-by: bongjin.lee <bongjin.lee@navercorp.com> Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com> Co-authored-by: Ji Chen <jimchen90@devfair0160.h2.fair> Co-authored-by: moto <855818+mthrok@users.noreply.github.com> Co-authored-by: Eli Uriegas <1700823+seemethere@users.noreply.github.com> Co-authored-by: lbjcom <lbjcom@gmail.com> Co-authored-by: jimchen90 <65626796+jimchen90@users.noreply.github.com> Co-authored-by: Artyom Astafurov <asta@fb.com> Co-authored-by: engineerchuan <engineerchuan@gmail.com>
pytorch · Jul 20, 2020 · f17ae39 · f17ae39
1 parent 56aa954
commit f17ae39
Show file tree

Hide file tree

Showing 55 changed files with 2,809 additions and 1,037 deletions.
diff --git a/.circleci/config.yml b/.circleci/config.yml
@@ -43,7 +43,7 @@ commands:
     steps:
       - run:
           name: Install cmake and pkg-config
-          command: HOMEBREW_NO_AUTO_UPDATE=1 brew install cmake pkg-config
+          command: HOMEBREW_NO_AUTO_UPDATE=1 brew install cmake pkg-config wget
           # Disable brew auto update which is very slow
 
 binary_common: &binary_common
@@ -533,6 +533,47 @@ jobs:
       - store_test_results:
           path: test-results
 
+  unittest_macos_cpu:
+    <<: *binary_common
+    macos:
+      xcode: "9.0"
+    resource_class: large
+    steps:
+      - checkout
+      - install_build_tools_macos
+      - attach_workspace:
+          at: third_party
+      - generate_cache_key
+      - restore_cache:
+
+          keys:
+            - env-v3-macos-{{ arch }}-py<< parameters.python_version >>-{{ checksum ".circleci/unittest/linux/scripts/environment.yml" }}-{{ checksum "third_party/CMakeLists.txt" }}-{{ checksum ".cachekey" }}
+
+      - run:
+          name: Setup
+          command: .circleci/unittest/linux/scripts/setup_env.sh
+      - save_cache:
+
+          key: env-v3-macos-{{ arch }}-py<< parameters.python_version >>-{{ checksum ".circleci/unittest/linux/scripts/environment.yml" }}-{{ checksum "third_party/CMakeLists.txt" }}-{{ checksum ".cachekey" }}
+
+          paths:
+            - conda
+            - env
+            - third_party/build
+            - third_party/install
+            - third_party/src
+      - run:
+          name: Install torchaudio
+          command: .circleci/unittest/linux/scripts/install.sh
+      - run:
+          name: Run tests
+          command: .circleci/unittest/linux/scripts/run_test.sh
+      - run:
+          name: Post process
+          command: .circleci/unittest/linux/scripts/post_process.sh
+      - store_test_results:
+          path: test-results
+
   stylecheck:
     <<: *binary_common
     docker:
@@ -744,6 +785,21 @@ workflows:
               only: /v[0-9]+(\.[0-9]+)*-rc[0-9]+/
           name: unittest_windows_gpu_py3.8
           python_version: '3.8'
+      - unittest_macos_cpu:
+          name: unittest_macos_cpu_py3.6
+          python_version: '3.6'
+          requires:
+          - download_third_parties_nix
+      - unittest_macos_cpu:
+          name: unittest_macos_cpu_py3.7
+          python_version: '3.7'
+          requires:
+          - download_third_parties_nix
+      - unittest_macos_cpu:
+          name: unittest_macos_cpu_py3.8
+          python_version: '3.8'
+          requires:
+          - download_third_parties_nix
   nightly:
     jobs:
       - circleci_consistency:

diff --git a/.circleci/config.yml.in b/.circleci/config.yml.in
@@ -43,7 +43,7 @@ commands:
     steps:
       - run:
           name: Install cmake and pkg-config
-          command: HOMEBREW_NO_AUTO_UPDATE=1 brew install cmake pkg-config
+          command: HOMEBREW_NO_AUTO_UPDATE=1 brew install cmake pkg-config wget
           # Disable brew auto update which is very slow
 
 binary_common: &binary_common
@@ -533,6 +533,47 @@ jobs:
       - store_test_results:
           path: test-results
 
+  unittest_macos_cpu:
+    <<: *binary_common
+    macos:
+      xcode: "9.0"
+    resource_class: large
+    steps:
+      - checkout
+      - install_build_tools_macos
+      - attach_workspace:
+          at: third_party
+      - generate_cache_key
+      - restore_cache:
+          {% raw %}
+          keys:
+            - env-v3-macos-{{ arch }}-py<< parameters.python_version >>-{{ checksum ".circleci/unittest/linux/scripts/environment.yml" }}-{{ checksum "third_party/CMakeLists.txt" }}-{{ checksum ".cachekey" }}
+          {% endraw %}
+      - run:
+          name: Setup
+          command: .circleci/unittest/linux/scripts/setup_env.sh
+      - save_cache:
+          {% raw %}
+          key: env-v3-macos-{{ arch }}-py<< parameters.python_version >>-{{ checksum ".circleci/unittest/linux/scripts/environment.yml" }}-{{ checksum "third_party/CMakeLists.txt" }}-{{ checksum ".cachekey" }}
+          {% endraw %}
+          paths:
+            - conda
+            - env
+            - third_party/build
+            - third_party/install
+            - third_party/src
+      - run:
+          name: Install torchaudio
+          command: .circleci/unittest/linux/scripts/install.sh
+      - run:
+          name: Run tests
+          command: .circleci/unittest/linux/scripts/run_test.sh
+      - run:
+          name: Post process
+          command: .circleci/unittest/linux/scripts/post_process.sh
+      - store_test_results:
+          path: test-results
+
   stylecheck:
     <<: *binary_common
     docker:

diff --git a/.circleci/regenerate.py b/.circleci/regenerate.py
@@ -136,8 +136,11 @@ def indent(indentation, data_list):
 def unittest_workflows(indentation=6):
     jobs = []
     jobs += build_download_job(None)
-    for os_type in ["linux", "windows"]:
+    for os_type in ["linux", "windows", "macos"]:
         for device_type in ["cpu", "gpu"]:
+            if os_type == "macos" and device_type == "gpu":
+                continue
+
             for i, python_version in enumerate(PYTHON_VERSIONS):
                 job = {
                     "name": f"unittest_{os_type}_{device_type}_py{python_version}",

diff --git a/.circleci/unittest/linux/scripts/environment.yml b/.circleci/unittest/linux/scripts/environment.yml
@@ -8,10 +8,10 @@ dependencies:
   - pytest-cov
   - codecov
   - librosa
+  - llvmlite==0.31  # See https://github.com/pytorch/audio/pull/766
   - pip
   - pip:
-    - clang-format
     - kaldi-io
     - scipy
     - parameterized
-    - numba==0.48
+    - numba==0.48  # See https://github.com/librosa/librosa/issues/1160
diff --git a/.circleci/unittest/linux/scripts/setup_env.sh b/.circleci/unittest/linux/scripts/setup_env.sh
@@ -14,10 +14,15 @@ env_dir="${root_dir}/env"
 
 cd "${root_dir}"
 
+case "$(uname -s)" in
+    Darwin*) os=MacOSX;;
+    *) os=Linux
+esac
+
 # 1. Install conda at ./conda
 if [ ! -d "${conda_dir}" ]; then
     printf "* Installing conda\n"
-    wget -O miniconda.sh http://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh
+    wget -O miniconda.sh "http://repo.continuum.io/miniconda/Miniconda3-latest-${os}-x86_64.sh"
     bash ./miniconda.sh -b -f -p "${conda_dir}"
 fi
 eval "$(${conda_dir}/bin/conda shell.bash hook)"
@@ -32,6 +37,9 @@ conda activate "${env_dir}"
 # 3. Install Conda dependencies
 printf "* Installing dependencies (except PyTorch)\n"
 conda env update --file "${this_dir}/environment.yml" --prune
+if [ "${os}" == Linux ] ; then
+    pip install clang-format
+fi
 
 # 4. Buld codecs
 mkdir -p third_party/build

diff --git a/.github/pytorch-probot.yml b/.github/pytorch-probot.yml
@@ -0,0 +1 @@
+tracking_issue: 736
diff --git a/test/README.md b/test/README.md
@@ -41,6 +41,58 @@ The following test modules are defined for corresponding `torchaudio` module/fun
 - [assets/kaldi](./assets/kaldi): Contains Kaldi format matrix files used in [./test_compliance_kaldi.py](./test_compliance_kaldi.py).
 - [compliance](./compliance): Scripts used to generate above Kaldi matrix files.
 
+### Waveforms for Testing Purposes
+
+When testing transforms we often need waveforms of specific type (ex: pure tone, noise, or voice), with specific bitrate (ex. 8 or 16 kHz) and number of channels (ex. mono, stereo). Below are some tips on how to construct waveforms and guidance around existing audio files.
+
+#### Load a Waveform from a File
+
+```python
+filepath = common_utils.get_asset_path('filename.wav')
+waveform, sample_rate = common_utils.load_wav(filepath)
+```
+
+*Note: Should you choose to contribute an audio file, please leave a comment in the issue or pull request, mentioning content source and licensing information. WAV files are preferred. Other formats should be used only when there is no alternative. (i.e. dataset implementation comes with hardcoded non-wav extension).*
+
+#### Pure Tone
+
+Code:
+
+```python
+waveform = common_utils.get_sinusoid(
+	frequency=300,
+	sample_rate=16000,
+	duration=1,  # seconds
+	n_channels=1,
+	dtype="float32",
+	device="cpu",
+)
+```
+
+#### Noise
+
+Code:
+
+```python
+tensor = common_utils.get_whitenoise()
+```
+
+Files:
+
+* `steam-train-whistle-daniel_simon.wav`
+
+#### Voice
+
+Files:
+
+* `CommonVoice/cv-corpus-4-2019-12-10/tt/clips/common_voice_tt_00000000.wav`
+* `LibriSpeech/dev-clean/1272/128104/1272-128104-0000.flac`
+* `LJSpeech-1.1/wavs/LJ001-0001.wav`
+* `SpeechCommands/speech_commands_v0.02/go/0a9f9af7_nohash_0.wav`
+* `VCTK-Corpus/wav48/p224/p224_002.wav`
+* `waves_yesno/0_1_0_1_0_1_1_0.wav`
+* `vad-go-stereo-44100.wav`
+* `vad-go-mono-32000.wav`
 
 ## Adding test
 

diff --git a/test/assets/CommonVoice/cv-corpus-4-2019-12-10/tt/clips/common_voice_tt_00000000.mp3 b/test/assets/CommonVoice/cv-corpus-4-2019-12-10/tt/clips/common_voice_tt_00000000.mp3
diff --git a/test/assets/CommonVoice/cv-corpus-4-2019-12-10/tt/clips/common_voice_tt_00000000.wav b/test/assets/CommonVoice/cv-corpus-4-2019-12-10/tt/clips/common_voice_tt_00000000.wav
diff --git a/test/assets/CommonVoice/cv-corpus-4-2019-12-10/tt/train.tsv b/test/assets/CommonVoice/cv-corpus-4-2019-12-10/tt/train.tsv
@@ -1,3 +1,3 @@
 client_id	path	sentence	up_votes	down_votes	age	gender	accent
-00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000	common_voice_tt_00000000.mp3	test.	1	0	thirties	female	
-00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001	common_voice_tt_00000000.mp3	test.	1	0	thirties	female	
+00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000	common_voice_tt_00000000.wav	test.	1	0	thirties	female	
+00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001	common_voice_tt_00000000.wav	test.	1	0	thirties	female	
diff --git a/test/assets/dtmf_30s_stereo.mp3 b/test/assets/dtmf_30s_stereo.mp3