feat:update dockerfile

.github/workflows/ci.yml

@@ -13,7 +13,7 @@ permissions:
   contents: read
 
 env:
-  NLTK_DATA: /home/notebook-user/nltk_data
+  NLTK_DATA: ${{ github.workspace }}/nltk_data
 
 jobs:
   setup:

.github/workflows/ingest-test-fixtures-update-pr.yml

@@ -14,7 +14,7 @@ jobs:
   setup:
     runs-on: ubuntu-latest-m
     env:
-      NLTK_DATA: /home/notebook-user/nltk_data
+      NLTK_DATA: ${{ github.workspace }}/nltk_data
     steps:
     - uses: actions/checkout@v3
     - uses: ./.github/actions/base-cache
@@ -24,7 +24,7 @@ jobs:
   setup_ingest:
     runs-on: ubuntu-latest
     env:
-      NLTK_DATA: /home/notebook-user/nltk_data
+      NLTK_DATA: ${{ github.workspace }}/nltk_data
     needs: [setup]
     steps:
       - uses: actions/checkout@v3

Dockerfile

@@ -9,8 +9,6 @@ COPY unstructured unstructured
 COPY test_unstructured test_unstructured
 COPY example-docs example-docs
 
-# Copy the downloaded NLTK data folder to your local environment.s
-COPY ./nltk_data /home/notebook-user/nltk_data
 
 RUN chown -R notebook-user:notebook-user /app && \
     apk add font-ubuntu git && \
@@ -22,7 +20,7 @@ USER notebook-user
 RUN find requirements/ -type f -name "*.txt" -exec pip3.11 install --no-cache-dir --user -r '{}' ';'
 
 # Command to check if NLTK data has been copied correctly
-RUN python3.11 -c "import nltk; print(nltk.data.find('tokenizers/punkt_tab'))" 
+RUN python3.11 -c "from unstructured.nlp.tokenize import copy_nltk_packages; copy_nltk_packages()"
 
 RUN python3.11 -c "from unstructured.partition.model_init import initialize; initialize()" && \
     python3.11 -c "from unstructured_inference.models.tables import UnstructuredTableTransformerModel; model = UnstructuredTableTransformerModel(); model.initialize('microsoft/table-transformer-structure-recognition')"