tfx

tfx copied to clipboard

• Have I specified the code to reproduce the issue (Yes, No): No
• Environment in which the code is executed (e.g., Local(Linux/MacOS/Windows), Interactive Notebook, Google Cloud, etc): Google Cloud
• TensorFlow version: 2.7.0
• TFX Version: > 1.5.0
• Python version: 3.7
• Python dependencies (from pip freeze output):

Describe the current behavior

When using 'example_gen_pb2.PayloadFormat.FORMAT_TF_SEQUENCE_EXAMPLE' payload_format in ImportExampleGen, transform component throws the following error:

AssertionError: read_as_raw_records is specified - must provide raw_record_column_name

The code used for ImportExampleGen is :

tfx.components.ImportExampleGen(input_base=data_path,input_config=example_gen_input_config, output_config=example_gen_output_config, payload_format=example_gen_pb2.PayloadFormat.FORMAT_TF_SEQUENCE_EXAMPLE)

Other info / logs

File "runner.py", line 138, in config_args = {'feature_config':"test", 'model_config':"test"} File "/opt/conda/lib/python3.7/site-packages/tfx/orchestration/portable/tfx_runner.py", line 116, in run return self.run_with_ir(pipeline_pb, run_options=run_options_pb) File "/opt/conda/lib/python3.7/site-packages/tfx/orchestration/local/local_dag_runner.py", line 109, in run_with_ir component_launcher.launch() File "/opt/conda/lib/python3.7/site-packages/tfx/orchestration/portable/launcher.py", line 545, in launch executor_output = self._run_executor(execution_info) File "/opt/conda/lib/python3.7/site-packages/tfx/orchestration/portable/launcher.py", line 420, in _run_executor executor_output = self._executor_operator.run_executor(execution_info) File "/opt/conda/lib/python3.7/site-packages/tfx/orchestration/portable/beam_executor_operator.py", line 98, in run_executor return python_executor_operator.run_with_executor(execution_info, executor) File "/opt/conda/lib/python3.7/site-packages/tfx/orchestration/portable/python_executor_operator.py", line 59, in run_with_executor execution_info.exec_properties) File "/opt/conda/lib/python3.7/site-packages/tfx/components/transform/executor.py", line 576, in Do TransformProcessor().Transform(label_inputs, label_outputs, status_file) File "/opt/conda/lib/python3.7/site-packages/tfx/components/transform/executor.py", line 1068, in Transform d.tfxio = self._CreateTFXIO(d, input_dataset_metadata.schema) File "/opt/conda/lib/python3.7/site-packages/tfx/components/transform/executor.py", line 1623, in _CreateTFXIO read_as_raw_records=read_as_raw_records) File "/opt/conda/lib/python3.7/site-packages/tfx/components/util/tfxio_utils.py", line 302, in make_tfxio 'read_as_raw_records is specified - ' AssertionError: read_as_raw_records is specified - must provide raw_record_column_name

Zohar, do you know if TF_SEQUENCE_EXAMPLE is supported by Transform?

It currently has basic support, to be extended some time soon. Currently, SequenceExamples are supported by providing serialized SequenceExamples as a single feature to the preprocessing_fn which can then be decoded and proceed by the user. Though this appears to be currently broken, let me take a look.

On Tue, Mar 8, 2022 at 1:35 AM Jiayi Zhao @.***> wrote:

Zohar, do you know if TF_SEQUENCE_EXAMPLE is supported by Transform?

— Reply to this email directly, view it on GitHub https://github.com/tensorflow/tfx/issues/4714#issuecomment-1061284444, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEGO5N6WQPKBB4G5YIMRRPLU62OF5ANCNFSM5PY6CAIQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were assigned.Message ID: @.***>