Merge branch 'master' into abbreviation_replacement_op

.github/workflows/main.yml

@@ -73,7 +73,7 @@ jobs:
 
       - name: Install Forte
         run: |
-          pip install --use-feature=in-tree-build --progress-bar off .[data_aug,ir,remote,audio_ext,stave,models,test,wikipedia,nlp,extractor]
+          pip install --progress-bar off .[data_aug,ir,remote,audio_ext,stave,models,test,wikipedia,nlp,extractor]
       - name: Install deep learning frameworks
         run: |
           pip install --progress-bar off torch==${{ matrix.torch-version }}
@@ -108,7 +108,7 @@ jobs:
             cd forte-wrappers
             for d in ${{ matrix.notebook-details.dep }}; do pip install "src/"$d; done
             cd ..
-            pip install --use-feature=in-tree-build --progress-bar off .[data_aug,ir,remote,audio_ext,stave,models,test,wikipedia,nlp,extractor]
+            pip install --progress-bar off .[data_aug,ir,remote,audio_ext,stave,models,test,wikipedia,nlp,extractor]
             coverage run -m pytest tests/forte/notebooks
           fi
   test_backbone:
@@ -132,7 +132,7 @@ jobs:
       - name: Test backbone Forte import test
         run: |
           # Try to install Forte backbone only and test basic imports.
-          pip install --use-feature=in-tree-build --progress-bar off .
+          pip install --progress-bar off .
           python tests/forte/forte_backbone_test.py
 
   test_modules:
@@ -168,7 +168,7 @@ jobs:
             ${{ runner.os }}-
       - name: install forte
         run: |
-          pip install --use-feature=in-tree-build --progress-bar off .
+          pip install --progress-bar off .
       - name: install dependencies
         run: |
           sudo apt-get install -y libsndfile1-dev
@@ -186,8 +186,8 @@ jobs:
           rm -rf stave
       - name: Test modules need extra packages
         run: |
-          pip install --use-feature=in-tree-build --progress-bar off .[test]
-          pip install --use-feature=in-tree-build --progress-bar off .[${{ matrix.modules.module }}]
+          pip install --progress-bar off .[test]
+          pip install --progress-bar off .[${{ matrix.modules.module }}]
           pip install --progress-bar off torch==${{ matrix.torch-version }}
           pip install --progress-bar off pytest==5.1.3
           pytest ${{ matrix.modules.test_file }}

examples/pipelines/README.md

@@ -2,9 +2,9 @@
 
 The examples in this folder
 
-- Build a pipeline using third-party tools like NLTK and StanfordNLP. 
+- Build a pipeline using third-party tools like NLTK and StanfordNLP.
 
-- Demonstrate integration of processors developed using third-party tools into a single NLP 
+- Demonstrate integration of processors developed using third-party tools into a single NLP
 pipeline.
 
 
@@ -18,7 +18,7 @@ pipeline.
 pip install forte.nltk
 ```
 
-- NER and SRL processors are built using texar-pytorch. Please follow the guidelines here 
+- NER and SRL processors are built using texar-pytorch. Please follow the guidelines here
 https://github.com/asyml/texar-pytorch#installation to install it.
 
 - To run `stanford_nlp_pipeline.py`, install Stanford NLP library using
@@ -38,29 +38,33 @@ For SRL,
 python download_models.py --model-name srl
 ```
 
+
 For NER,
 ```bash
 python download_models.py --model-name ner
 ```
+**Note**: the hyperlink for the ner model has been outdated and no longer valid. If you want to try out the pipeline, this [training example](https://github.com/petuum/composing_information_system/blob/main/training.md) have models and datasets to all the valid hyperlinks.
+
 
-**Note**: The above script will save the model in `resources/`. Use `--path` option to save the 
+
+**Note**: The above script will save the model in `resources/`. Use `--path` option to save the
 model into a different directory.
 
 ## Running the pipeline
 
 `process_string_example.py` and `process_dataset_example.py` build the following pipeline
 
-Reader -> NLTKSentenceSegmenter -> NLTKWordTokenizer -> NLTKPOSTagger -> NER Predictor -> 
+Reader -> NLTKSentenceSegmenter -> NLTKWordTokenizer -> NLTKPOSTagger -> NER Predictor ->
 SRL Predictor
 
-The configuration for NER and SRL Predictor are read from `config.yml` file and the processors are 
-initialized during the pipeline creation. To see the pipeline in action, run 
+The configuration for NER and SRL Predictor are read from `config.yml` file and the processors are
+initialized during the pipeline creation. To see the pipeline in action, run
 
 ```bash
 python pipeline_string_example.py
 ```
 
-In `process_dataset_example.py`, we show the use of `process_dataset()` method of our pipeline 
+In `process_dataset_example.py`, we show the use of `process_dataset()` method of our pipeline
 which is used to read text files from a directory as data packs. To run this example,
 
 ```bash
@@ -77,4 +81,4 @@ We run this pipeline on an English as well as French sentence. To see this actio
 
 ```bash
 python stanford_nlp_pipeline.py
-``` 
+```