Vitis AI GPU Docker(TensorFlow 1.x) vai_q_tensorflow command errors and thier solution

[TeppeiT]

Hi

In the Vitis AI 3.0 repository, I built the GPU Docker for TensorFlow 1.x and after launching it,
I got the following SyntaxError when running the vai_q_tensorflow command.

It was a SyntaxError and we solved it by fixing some source code, so please consider fixing it if possible.

・Error

vitis-ai-user@xxx:/workspace$ vai_q_tensorflow --help
Traceback (most recent call last):
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/bin/vai_q_tensorflow", line 7, in <module>
    from vai_q_tensorflow.python.decent_q import run_main
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/lib/python3.6/site-packages/vai_q_tensorflow/__init__.py", line 38, in <module>
    from vai_q_tensorflow.python.decent_q import inspect
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/lib/python3.6/site-packages/vai_q_tensorflow/python/decent_q.py", line 91, in <module>
    from progressbar import ProgressBar
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/lib/python3.6/site-packages/progressbar/__init__.py", line 3, in <module>
    from .utils import (
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/lib/python3.6/site-packages/progressbar/utils.py", line 9, in <module>
    from python_utils.time import timedelta_to_seconds, epoch, format_time
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/lib/python3.6/site-packages/python_utils/__init__.py", line 1, in <module>
    from . import (
  File "/opt/vitis_ai/conda/envs/vitis-ai-tensorflow/lib/python3.6/site-packages/python_utils/converters.py", line 89
    if match := regexp.search(input_):
              ^
SyntaxError: invalid syntax

I changed “:=” to “==” in lines 89 and 170 of converters.py and the error was resolved.

・L89

 87     try:
 88         if regexp and input_:
 89             if match == regexp.search(input_):
 90             #if match := regexp.search(input_):
 91                 input_ = match.groups()[-1]

・L170

168     try:
169         if regexp:
170             if match == regexp.search(input_):
171             #if match := regexp.search(input_):
172                 input_ = match.group(1)

・after chenged

vitis-ai-user@xxx:/workspace$ vai_q_tensorflow --help
usage:
    usage: vai_q_tensorflow command [Options]

    examples:
      show help       : vai_q_tensorflow --help
      quantize a model: vai_q_tensorflow quantize --input_frozen_graph frozen_graph.pb --input_nodes xxx --output_nodes yyy --input_shapes zzz --input_fn module.calib_input
      inspect a model : vai_q_tensorflow inspect --input_frozen_graph frozen_graph.pb
      dump quantized model : vai_q_tensorflow dump --input_frozen_graph quantize_results/quantize_eval_model.pb --input_fn module.dump_input
      update old quantized model: vai_q_tensorflow update --input_frozen_graph quantize_results/quantize_eval_model.pb


Xilinx's Quantization ToolsVai_q_tensorflow 3.0.0-dev-gpu build for Tensorflow
1.15.2 git version d2e041a

positional arguments:
  {quantize,inspect,dump,deploy,update}
                        Specify a command for vai_q_tensorflow

optional arguments:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  --mode {frozen,train,eval}