yeandy commented on code in PR #22131: URL: https://github.com/apache/beam/pull/22131#discussion_r920470464
########## sdks/python/apache_beam/examples/inference/tensorrt_object_detection.py: ########## @@ -0,0 +1,238 @@ +# +# Licensed to the Apache Software Foundation (ASF) under one or more +# contributor license agreements. See the NOTICE file distributed with +# this work for additional information regarding copyright ownership. +# The ASF licenses this file to You under the Apache License, Version 2.0 +# (the "License"); you may not use this file except in compliance with +# the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# + +"""A pipeline that uses RunInference API to perform object detection with TensorRT.""" + +import argparse +import io +import numpy as np +import os +from PIL import Image +from typing import Iterable, Optional, Tuple + +import apache_beam as beam +from apache_beam.io.filesystems import FileSystems +from apache_beam.ml.inference.base import (KeyedModelHandler, PredictionResult, + RunInference) +from apache_beam.ml.inference.tensorrt_inference import \ + TensorRTEngineHandlerNumPy +from apache_beam.options.pipeline_options import PipelineOptions, SetupOptions + +COCO_OBJ_DET_CLASSES = [ + 'person', + 'bicycle', + 'car', + 'motorcycle', + 'airplane', + 'bus', + 'train', + 'truck', + 'boat', + 'traffic light', + 'fire hydrant', + 'street sign', + 'stop sign', + 'parking meter', + 'bench', + 'bird', + 'cat', + 'dog', + 'horse', + 'sheep', + 'cow', + 'elephant', + 'bear', + 'zebra', + 'giraffe', + 'hat', + 'backpack', + 'umbrella', + 'shoe', + 'eye glasses', + 'handbag', + 'tie', + 'suitcase', + 'frisbee', + 'skis', + 'snowboard', + 'sports ball', + 'kite', + 'baseball bat', + 'baseball glove', + 'skateboard', + 'surfboard', + 'tennis racket', + 'bottle', + 'plate', + 'wine glass', + 'cup', + 'fork', + 'knife', + 'spoon', + 'bowl', + 'banana', + 'apple', + 'sandwich', + 'orange', + 'broccoli', + 'carrot', + 'hot dog', + 'pizza', + 'donut', + 'cake', + 'chair', + 'couch', + 'potted plant', + 'bed', + 'mirror', + 'dining table', + 'window', + 'desk', + 'toilet', + 'door', + 'tv', + 'laptop', + 'mouse', + 'remote', + 'keyboard', + 'cell phone', + 'microwave', + 'oven', + 'toaster', + 'sink', + 'refrigerator', + 'blender', + 'book', + 'clock', + 'vase', + 'scissors', + 'teddy bear', + 'hair drier', + 'toothbrush', + 'hair brush', +] + + +def attach_im_size_to_key(x): + width, height = x[1].size + return ((x[0], width, height), x[1]) + + +def read_image(image_file_name: str, + path_to_dir: Optional[str] = None) -> Tuple[str, Image.Image]: + if path_to_dir is not None: + image_file_name = os.path.join(path_to_dir, image_file_name) + with FileSystems().open(image_file_name, 'r') as file: + data = Image.open(io.BytesIO(file.read())).convert('RGB') + return image_file_name, data + + +def preprocess_image(image: Image.Image) -> np.ndarray: + ssd_mobilenet_v2_320x320_input_dims = (300, 300) + image = image.resize(ssd_mobilenet_v2_320x320_input_dims, + resample=Image.Resampling.BILINEAR) + image = np.expand_dims(np.asarray(image, dtype=np.float32), axis=0) + return image + + +class PostProcessor(beam.DoFn): + def process(self, element: Tuple[str, PredictionResult]) -> Iterable[str]: + key, prediction_result = element + filename, im_width, im_height = key + nums = prediction_result.inference[0] + boxes = prediction_result.inference[1] + scores = prediction_result.inference[2] + classes = prediction_result.inference[3] + detections = [] + for i in range(int(nums[0])): + detections.append({ + 'ymin': str(boxes[i][0] * im_height), + 'xmin': str(boxes[i][1] * im_width), + 'ymax': str(boxes[i][2] * im_height), + 'xmax': str(boxes[i][3] * im_width), + 'score': str(scores[i]), + 'class': COCO_OBJ_DET_CLASSES[int(classes[i])] + }) + yield filename + ',' + str(detections) + +def parse_known_args(argv): + """Parses args for the workflow.""" + parser = argparse.ArgumentParser() + parser.add_argument( + '--input', + dest='input', + required=True, + help='Path to the text file containing image names.') + parser.add_argument( + '--output', + dest='output', + required=True, + help='Path where to save output predictions.' + ' text file.') + parser.add_argument( + '--engine_path', + dest='engine_path', + required=True, + help='Path to the pre-built TFOD ssd_mobilenet_v2_320x320_coco17_tpu-8' + 'TensorRT engine.') + parser.add_argument( + '--images_dir', + default=None, + help='Path to the directory where images are stored.' + 'Not required if image names in the input file have absolute path.') + return parser.parse_known_args(argv) + + +def run(argv=None, save_main_session=True): + """ + Args: + argv: Command line arguments defined for this example. + """ + known_args, pipeline_args = parse_known_args(argv) + pipeline_options = PipelineOptions(pipeline_args) + pipeline_options.view_as(SetupOptions).save_main_session = save_main_session + + engine_handler = KeyedModelHandler( + TensorRTEngineHandlerNumPy( + min_batch_size=1, max_batch_size=1, + engine_path=known_args.engine_path)) + + with beam.Pipeline(options=pipeline_options) as p: + filename_value_pair = ( + p + | 'ReadImageNames' >> beam.io.ReadFromText( + known_args.input, skip_header_lines=0) + | 'ReadImageData' >> beam.Map( + lambda image_name: read_image( + image_file_name=image_name, path_to_dir=known_args.images_dir)) + | 'AttachImageSizeToKey' >> beam.Map(attach_im_size_to_key) + | 'PreprocessImages' >> beam.MapTuple( + lambda file_name, data: (file_name, preprocess_image(data)))) + predictions = ( + filename_value_pair + | 'PyTorchRunInference' >> RunInference(engine_handler) + | 'ProcessOutput' >> beam.ParDo(PostProcessor())) + + if known_args.output: + predictions | "WriteOutputToGCS" >> beam.io.WriteToText( + known_args.output, + shard_name_template='', + append_trailing_newlines=True) + Review Comment: Originally, the output wasn't required, which is why we had the if statement. We then changed it to be required, but forgot to take out the if statement. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
