TFX Trainer component error
Aditya-171 opened this issue · 5 comments
While running the following pipeline, ->(https://github.com/tensorflow/tfx/blob/master/tfx/examples/penguin/penguin_pipeline_local.py)
Module file being used -> tfx/tfx/examples/penguin/penguin_utils_keras.py at master · tensorflow/tfx (github.com)
I am facing error in Trainer Component as
Failed to Convert elements of SparceTensor(indices=Tensor(“DeserilizeSparse_121:0”, shape=(None,2) dtype=int64), values=Tensor(" DeserilizeSparse_121:1", shape=(None, ), dtype=int64), dense_type=Tensor("Stack_121:0 ",shape=(2, ), dtype=int64) ) to Tensor
Getting the error as model.fit( )
Slight changes in the make_keras_model() give me error as
ValueError: Missing Data for input "dense_4_input". You passed a data dictionary with keys[, , , ,] . Expected the following keys:[dense_4_input ].
I am unable to figure out this issue
I am able to run the penguin example on local runner successfully without any changes by following the instructions in example. Please find below the exact steps below and make sure bazel in installed in you environment. Ref: Screenshot below
# create python 3.10 environment
# activate environment
# install bazel
# clone tfx repository
git clone https://github.com/tensorflow/tfx ~/tfx-source
# cd to tfx folder
cd tfx
# install tfx nightly latest
export TFX_DEPENDENCY_SELECTOR=NIGHTLY
pip install -e . --extra-index-url https://pypi-nightly.tensorflow.org/simple
pip install -e .[kfp] --extra-index-url https://pypi-nightly.tensorflow.org/simple
# copy penguin folder from tfx folder to local
cp -r ~/tfx-source/tfx/examples/penguin ~/
# run penguin example
python ~/penguin/penguin_pipeline_local.py
This issue has been marked stale because it has no recent activity since 7 days. It will be closed if no further activity occurs. Thank you.
This issue was closed due to lack of activity after being marked stale for past 7 days.