Merge branch 'dialogue_state_tracking_refactor' of https://github.com…

…/NVIDIA/NeMo into main
NVIDIA · Jan 28, 2022 · 7c12ab6 · 7c12ab6
2 parents 39dbec3 + 4bd1cc1
commit 7c12ab6
Show file tree

Hide file tree

Showing 334 changed files with 18,475 additions and 7,577 deletions.
diff --git a/.github/workflows/changelog-build.yml b/.github/workflows/changelog-build.yml
@@ -0,0 +1,47 @@
+name: 'Changelog Build (Release)'
+
+on:
+  push:
+    tags:
+      - '*'
+
+jobs:
+  changelog:
+    if: startsWith(github.ref, 'refs/tags/')
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/[email protected]
+        with:
+          fetch-depth: 0 # Required due to the way Git works, without it this action won't be able to find any or the correct tags
+
+      - name: Get Previous tag
+        id: previous_tag
+        # git for-each-ref --sort=-creatordate --format '%(refname)' refs/tags ==> refs/tags/vX.Y.Z in descending order of date
+        # awk 'FNR == 2 {print substr($1, 11, length($1))}') ==> Selects the 2nd tag from the list, then strips the /refs/tags/ part of the tag
+        # set-output name=tag_name:: ==> Takes the clean tag vX.Y.Z and sets it to steps.previous_tag.outputs.tag_name
+        run: |
+          echo "::set-output name=tag_name::$(git for-each-ref --sort=-creatordate --format '%(refname)' refs/tags | awk 'FNR == 2 {print substr($1, 11, length($1))}')"
+          echo ${{ steps.previous_tag.outputs.tag_name }}
+
+      - name: Build Changelog
+        id: github_tag
+        uses: mikepenz/[email protected]
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        with:
+          # Configuration file is setup with filters for domains
+          # owner:repo must point to current repo
+          # fromTag: Auto resolved from historical tag order (previous tag compared to current tag)
+          # toTag: Current tag reference
+          configuration: ".github/workflows/config/changelog-config.json"
+          owner: "NVIDIA"
+          repo: "NeMo"
+          ignorePreReleases: "false"
+          failOnError: "false"
+          fromTag: ${{ steps.previous_tag.outputs.tag_name }}
+          toTag: ${{ github.ref_name }}
+
+      - name: Print Changelog
+        run: |
+          echo "${{steps.github_tag.outputs.changelog}}"
+          echo "--- DONE ---"
diff --git a/.github/workflows/config/changelog-config.json b/.github/workflows/config/changelog-config.json
@@ -0,0 +1,168 @@
+{
+    "categories": [
+      {
+        "title": "## ASR",
+        "labels": ["asr"]
+      },
+      {
+        "title": "## TTS",
+        "labels": ["tts"]
+      },
+      {
+        "title": "## NLP / NMT",
+        "labels": ["nlp", "nmt", "megatron"]
+      },
+      {
+        "title": "## Text Normalization / Inverse Text Normalization",
+        "labels": ["tn", "itn"]
+      },
+      {
+        "title": "## NeMo Tools",
+        "labels": ["tools"]
+      },
+      {
+        "title": "## Export",
+        "labels": ["export"]
+      },
+      {
+        "title": "## Documentation",
+        "labels": ["docs"]
+      },
+      {
+        "title": "## Bugfixes",
+        "labels": ["bug"]
+      }
+    ],
+    "ignore_labels": [
+      "ignore"
+    ],
+    "sort": "ASC",
+    "template": "${{CHANGELOG}}\nUncategorized:\n${{UNCATEGORIZED}}\n\n",
+    "pr_template": "- ${{TITLE}} :: PR: #${{NUMBER}}",
+    "empty_template": "${{OWNER}}\n${{REPO}}\n${{FROM_TAG}}\n${{TO_TAG}}",
+    "label_extractor": [
+      {
+        "pattern": "(.*tts.*)|(.*g2p.*)",
+        "target": "tts",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*tts.*)|(.*g2p.*)",
+        "target": "tts",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*asr.*)|(.*ctc.*)|(.*rnnt.*)|(.*transducer.*)|(.*dali.*)",
+        "target": "asr",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*asr.*)|(.*ctc.*)|(.*rnnt.*)|(.*transducer.*)|(.*dali.*)",
+        "target": "asr",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*nlp.*)|(.*punctuation.*)|(.*capitalization.*)|(.*entity.*)|(.*glue.*)|(.*entity.*)|(.*retrieval.*)|(.*entity.*)|(.*intent.*)|(.*slot.*)|(.*entity.*)|(.*language.*)|(.*qa.*)|(.*token class.*)|(.*text class.*)",
+        "target": "nlp",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*nlp.*)|(.*punctuation.*)|(.*capitalization.*)|(.*entity.*)|(.*glue.*)|(.*entity.*)|(.*retrieval.*)|(.*entity.*)|(.*intent.*)|(.*slot.*)|(.*entity.*)|(.*language.*)|(.*qa.*)|(.*token class.*)|(.*text class.*)",
+        "target": "nlp",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*nmt.*)|(.*bignlp.*)|(.*megatron.*)|(.*machine.*)|(.*translation.*)|(.*gpt.*)",
+        "target": "nmt",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*nmt.*)|(.*bignlp.*)|(.*megatron.*)|(.*machine.*)|(.*translation.*)|(.*gpt.*)",
+        "target": "nmt",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*tn.*)|(.*itn.*)|(.*text norm.*)",
+        "target": "tn",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*tn.*)|(.*itn.*)|(.*text norm.*)",
+        "target": "tn",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*sde.*)|(.*ctc segment.*)",
+        "target": "tools",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*sde.*)|(.*ctc segment.*)",
+        "target": "tools",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*trt.*)|(.*onnx.*)",
+        "target": "export",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*trt.*)|(.*onnx.*)",
+        "target": "export",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*doc.*)",
+        "target": "docs",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*doc.*)",
+        "target": "docs",
+        "flags": "gimu",
+        "on_property": "body"
+      },
+      {
+        "pattern": "(.*bug.*)|(.*bugfix.*)|(.*fix.*)|(.*patch.*)",
+        "target": "bug",
+        "flags": "gimu",
+        "on_property": "title"
+      },
+      {
+        "pattern": "(.*bug.*)|(.*bugfix.*)|(.*fix.*)|(.*patch.*)",
+        "target": "bug",
+        "flags": "gimu",
+        "on_property": "body"
+      }
+    ],
+    "duplicate_filter": {
+      "pattern": ".+",
+      "on_property": "title",
+      "method": "match"
+    },
+    "transformers": [
+    ],
+    "max_tags_to_fetch": 100,
+    "max_pull_requests": 500,
+    "max_back_track_time_days": 365,
+    "exclude_merge_branches": [
+    ],
+    "tag_resolver": {
+      "method": "semver"
+    }
+}
diff --git a/Dockerfile b/Dockerfile
@@ -14,7 +14,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-ARG BASE_IMAGE=nvcr.io/nvidia/pytorch:21.11-py3
+ARG BASE_IMAGE=nvcr.io/nvidia/pytorch:21.12-py3
 
 
 # build an image that includes only the nemo dependencies, ensures that dependencies
@@ -101,5 +101,4 @@ COPY tutorials /workspace/nemo/tutorials
 # COPY README.rst LICENSE /workspace/nemo/
 
 RUN printf "#!/bin/bash\njupyter lab --no-browser --allow-root --ip=0.0.0.0" >> start-jupyter.sh && \
-    chmod +x start-jupyter.sh
-
+    chmod +x start-jupyter.sh