diff --git a/requirements.txt b/requirements.txt
index fc710cc0..d2534489 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -8,7 +8,7 @@ psutil==5.8.0
 # frameworks
 ray==1.9.2
 lightgbm-ray==0.1.2
-lightgbm==3.3.0
+lightgbm==3.3.1
 treelite==2.1.0
 treelite_runtime==2.1.0
 flaml==0.9.6
diff --git a/src/scripts/data_processing/generate_data/generate.py b/src/scripts/data_processing/generate_data/generate.py
index 50af07ee..ba5445d7 100644
--- a/src/scripts/data_processing/generate_data/generate.py
+++ b/src/scripts/data_processing/generate_data/generate.py
@@ -252,6 +252,7 @@ def run(self, args, logger, metrics_logger, unknown_args):
         os.makedirs(args.output_train, exist_ok=True)
         os.makedirs(args.output_test, exist_ok=True)
         os.makedirs(args.output_inference, exist_ok=True)
+        os.makedirs(args.external_header, exist_ok=True)
 
 
         # transform delimiter
diff --git a/src/scripts/data_processing/generate_data/spec.yaml b/src/scripts/data_processing/generate_data/spec.yaml
index 33433b10..339a8a8c 100644
--- a/src/scripts/data_processing/generate_data/spec.yaml
+++ b/src/scripts/data_processing/generate_data/spec.yaml
@@ -111,7 +111,7 @@ outputs:
     type: AnyDirectory
   external_header:
     type: AnyDirectory
-
+    
 command: >-
   python generate.py
   --type {inputs.learning_task}
@@ -135,7 +135,7 @@ command: >-
   [--custom_properties {inputs.custom_properties}]
   [--docs_per_query {inputs.docs_per_query}]
   [--n_label_classes {inputs.n_label_classes}]
-  
+
 environment:
   conda:
     # conda file path is resolved after additional includes
diff --git a/src/scripts/training/lightgbm_python/default.dockerfile b/src/scripts/training/lightgbm_python/default.dockerfile
index 1a1ae441..9bf0fcd5 100644
--- a/src/scripts/training/lightgbm_python/default.dockerfile
+++ b/src/scripts/training/lightgbm_python/default.dockerfile
@@ -1,10 +1,29 @@
 FROM mcr.microsoft.com/azureml/openmpi4.1.0-ubuntu20.04:latest
-LABEL lightgbmbenchmark.linux.cpu.mpi.pip.version="3.3.0/20210118.1"
+LABEL lightgbmbenchmark.linux.cpu.mpi.pip.version="3.3.1/20211210.1"
+
+# https://github.com/microsoft/lightgbm-transform/blob/main/docs/Installation-Guide.rst
+# Install CMake, gcc, g++, boost.
+RUN apt-get update && apt-get -y upgrade && DEBIAN_FRONTEND="noninteractive" apt-get install -y libboost-all-dev gcc g++ wget cmake git curl libtinfo5
+
+# Install LLVM with RTTI feature.
+WORKDIR /root
+RUN wget https://github.com/llvm/llvm-project/archive/refs/tags/llvmorg-3.5.1.tar.gz && tar zxvf llvmorg-3.5.1.tar.gz
+WORKDIR /root/llvm-project-llvmorg-3.5.1/llvm
+RUN mkdir build && cd build && cmake -DLLVM_REQUIRES_RTTI=1 .. && make -j4 && make install
+
+# Install bond.
+WORKDIR /root
+RUN git clone --recursive https://github.com/microsoft/bond.git
+RUN DEBIAN_FRONTEND="noninteractive" apt-get install -y clang zlib1g-dev
+RUN curl -sSL https://get.haskellstack.org/ | sh
+WORKDIR /root/bond
+RUN mkdir build && cd build && cmake -DBOND_ENABLE_GRPC=FALSE .. && make -j4 && make install
+
 
 # Those arguments will NOT be used by AzureML
 # they are here just to allow for lightgbm-benchmark build to actually check
 # dockerfiles in a PR against their actual branch
-ARG lightgbm_version="3.3.0"
+ARG lightgbm_version="3.3.1"
 
 ENV AZUREML_CONDA_ENVIRONMENT_PATH /azureml-envs/lightgbm
 
@@ -27,7 +46,13 @@ RUN HOROVOD_WITH_TENSORFLOW=1 \
                 'azureml-mlflow==1.35.0' \
                 'azureml-telemetry==1.35.0' \
                 'mpi4py==3.1.1' \
-                lightgbm==${lightgbm_version}
+                'omegaconf'
+
+# install lightgbm with mpi
+RUN pip install --upgrade pip setuptools wheel && \
+    pip install 'cmake==3.21.0' && \
+    pip install lightgbm==${lightgbm_version} --install-option=--mpi &&\
+    pip install lightgbm-transform==${lightgbm_version}
 
 # This is needed for mpi to locate libpython
 ENV LD_LIBRARY_PATH $AZUREML_CONDA_ENVIRONMENT_PATH/lib:$LD_LIBRARY_PATH
diff --git a/src/scripts/training/lightgbm_python/spec.yaml b/src/scripts/training/lightgbm_python/spec.yaml
index f8805df3..c3ee88fc 100644
--- a/src/scripts/training/lightgbm_python/spec.yaml
+++ b/src/scripts/training/lightgbm_python/spec.yaml
@@ -30,6 +30,10 @@ inputs:
     type: AnyDirectory
     description: Directory to the testing data
     optional: false
+  parser_config_file:
+    type: AnyDirectory
+    description: directory to the transform parser config
+    optional: true
   construct:
     type: Boolean
     description: "Use lazy intialization during data loading phase (both train and test datasets)"
@@ -167,6 +171,7 @@ launcher:
     python train.py
     --train {inputs.train}
     --test {inputs.test}
+    [--parser_config_file {inputs.parser_config_file}]
     --construct {inputs.construct}
     --header {inputs.header}
     --label_column {inputs.label_column}
diff --git a/src/scripts/training/lightgbm_python/train.py b/src/scripts/training/lightgbm_python/train.py
index c2e3de2a..fbd6e31c 100644
--- a/src/scripts/training/lightgbm_python/train.py
+++ b/src/scripts/training/lightgbm_python/train.py
@@ -63,6 +63,7 @@ def get_arg_parser(cls, parser=None):
         group_i.add_argument("--header", required=False, default=False, type=strtobool)
         group_i.add_argument("--label_column", required=False, default="0", type=str)
         group_i.add_argument("--group_column", required=False, default=None, type=str)
+        group_i.add_argument("--parser_config_file", required=False, default=None, type=str, help="transfomr parser config location (file path)")
 
         group_o = parser.add_argument_group(f"Outputs [{__name__}:{cls.__name__}]")
         group_o.add_argument("--export_model",