[PyT/TF/TF2] remove install html2text requirement for NLP models

Author: sharathts

PyTorch/LanguageModeling/BERT/Dockerfile

@@ -33,7 +33,7 @@ RUN pip install /workspace/install/python/tensorrtserver*.whl
 WORKDIR /workspace/bert
 RUN pip install --upgrade --no-cache-dir pip \
  && pip install --no-cache-dir \
- tqdm boto3 requests six ipdb h5py html2text nltk progressbar onnxruntime \
+ tqdm boto3 requests six ipdb h5py nltk progressbar onnxruntime \
  git+https://github.com/NVIDIA/dllogger wget
 
 RUN apt-get install -y iputils-ping

PyTorch/LanguageModeling/BERT/requirements.txt

@@ -8,7 +8,6 @@ six
 ipdb
 #Data processing
 h5py
-html2text
 nltk
 progressbar
 #Others

TensorFlow/LanguageModeling/BERT/Dockerfile

@@ -4,7 +4,7 @@ FROM ${FROM_IMAGE_NAME}
 
 RUN apt-get update && apt-get install -y pbzip2 pv bzip2 libcurl4 curl libb64-dev
 RUN pip install --upgrade pip
-RUN pip install toposort networkx pytest nltk tqdm html2text progressbar pynvml
+RUN pip install toposort networkx pytest nltk tqdm progressbar pynvml
 RUN pip --no-cache-dir --no-cache install git+https://github.com/NVIDIA/dllogger wget
 
 WORKDIR /workspace

TensorFlow/LanguageModeling/BERT/requirements.txt

@@ -5,5 +5,4 @@ networkx
 pytest
 nltk
 tqdm
-html2text
 progressbar

TensorFlow2/LanguageModeling/BERT/Dockerfile

@@ -41,7 +41,6 @@ RUN pip3 install \
   tensorflow_hub \
   pynvml \
   wget \
-  html2text \
   progressbar \
   git+https://github.com/NVIDIA/dllogger
 

TensorFlow2/LanguageModeling/BERT/run.sub

@@ -67,7 +67,6 @@ PIP_CMD="pip3 install \
   tensorflow_hub \
   pynvml \
   wget \
-  html2text \
   progressbar \
   git+https://github.com/NVIDIA/dllogger"
 

TensorFlow2/LanguageModeling/ELECTRA/Dockerfile

@@ -24,7 +24,7 @@ RUN git clone https://github.com/soskek/bookcorpus.git
 WORKDIR /workspace/electra
 RUN pip install --upgrade --no-cache-dir pip \
  && pip install --no-cache-dir \
- tqdm boto3 requests six ipdb h5py html2text nltk progressbar filelock tokenizers==0.7.0 \
+ tqdm boto3 requests six ipdb h5py nltk progressbar filelock tokenizers==0.7.0 \
  git+https://github.com/NVIDIA/dllogger \
  nvidia-ml-py3==7.352.0
 

TensorFlow2/LanguageModeling/ELECTRA/scripts/docker/launch.sh

@@ -26,4 +26,4 @@ docker run -it --rm \
   --privileged \
   -e LD_LIBRARY_PATH='/workspace/install/lib/' \
   -v $PWD:/workspace/electra \
-  electra $CMD
+  electra $CMD