AnandInguva commented on code in PR #23456: URL: https://github.com/apache/beam/pull/23456#discussion_r1001107826
########## sdks/python/apache_beam/examples/inference/tfx_bsl/tensorflow_image_classification.py: ########## @@ -16,12 +16,12 @@ # """ -A pipeline tha uses TFX RunInference API to perform Image classification. +A pipeline that uses TFX RunInference API to perform image classification. Please look at https://github.com/tensorflow/tfx-bsl/tree/master/tfx_bsl/beam. -Note: For the Tensorflow Model, it needs to be updated with a @tf.function - Signature to accept bytes as inputs and should have logic to decode - bytes to data which should be acceptable by the tensorflow model. +Note: A Tensorflow Model needs to be updated with a @tf.function + Signature in order to accept bytes as inputs, and should have logic to decode Review Comment: We can use any TF Model. this is one example where we load the TF model with modified signature to use the RUnInference transform on the images. We don't update the signature as part of this example. it is done prior to running this pipeline and saved in a path accessible by the pipeline -- 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]
