AnandInguva commented on code in PR #17462: URL: https://github.com/apache/beam/pull/17462#discussion_r889210162
########## sdks/python/apache_beam/examples/inference/pytorch_image_classification.py: ########## @@ -0,0 +1,146 @@ +# +# 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. +# + +"""Pipeline that uses RunInference API to perform classification task on imagenet dataset""" # pylint: disable=line-too-long + +import argparse +import io +import os +from functools import partial +from typing import Any +from typing import Iterable +from typing import Tuple +from typing import Union + +import apache_beam as beam +import torch +from apache_beam.io.filesystems import FileSystems +from apache_beam.ml.inference.api import PredictionResult +from apache_beam.ml.inference.api import RunInference +from apache_beam.ml.inference.pytorch_inference import PytorchModelLoader +from apache_beam.options.pipeline_options import PipelineOptions +from apache_beam.options.pipeline_options import SetupOptions +from PIL import Image +from torchvision import transforms +from torchvision.models.mobilenetv2 import MobileNetV2 + + +def read_image(image_file_name: str, + path_to_dir: 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(data: Image) -> torch.Tensor: + image_size = (224, 224) + # to use models in torch with imagenet weights, + # normalize the images using the below values. + # ref: https://pytorch.org/vision/stable/models.html# + normalize = transforms.Normalize( + mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) + transform = transforms.Compose([ + transforms.Resize(image_size), + transforms.ToTensor(), + normalize, + ]) + return transform(data) + + +class PostProcessor(beam.DoFn): + def process( + self, element: Union[PredictionResult, Tuple[Any, PredictionResult]] Review Comment: 1 seems to be simple for this example and it works. Thanks. I will commit that change -- 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]
