# TFX-Addons **URL:** https://discuss.ai.google.dev/c/tensorflow/special-interest-groups/tfx-addons/54.md [Latest](https://discuss.ai.google.dev/latest.md) · [Categories](https://discuss.ai.google.dev/categories.md) · [Tags](https://discuss.ai.google.dev/tags.md) --- ## [About the SIG TFX-Addons category](https://discuss.ai.google.dev/t/about-the-sig-tfx-addons-category/26578)
**Author:** [@kristen](https://discuss.ai.google.dev/u/kristen)\ **Replies:** 3\ **Last updated:** [May 23, 2021, 9:54pm UTC](https://discuss.ai.google.dev/t/about-the-sig-tfx-addons-category/26578 "2021-05-23T21:54:46Z")
SIG TFX-Addons facilitates community contributions of community-supported software and tools which can be used on top of core TFX. To contribute to TFX-Addons, join the conversation at our monthly meeting. --- ## [MacOS M2 upgrade Sonoma 14.0 can not train model with tensorflow and karas](https://discuss.ai.google.dev/t/macos-m2-upgrade-sonoma-14-0-can-not-train-model-with-tensorflow-and-karas/32287)
**Author:** [@Cloris\_Yin](https://discuss.ai.google.dev/u/Cloris_Yin)\ **Replies:** 1\ **Last updated:** [July 29, 2025, 7:39am UTC](https://discuss.ai.google.dev/t/macos-m2-upgrade-sonoma-14-0-can-not-train-model-with-tensorflow-and-karas/32287 "2025-07-29T07:39:49Z")
I can train a yolov3 at MacOS M2 ventura with tensorflow-macos=2.9.0(karas=2.9.0) and tensorflow-mental=0.5. But when I upgrade the system to Sonoma14.0. I can not train model with below error.’ \`MPSKernel MTLComputePi… --- ## [How to handle OOM at TFX Transform component?](https://discuss.ai.google.dev/t/how-to-handle-oom-at-tfx-transform-component/31611)
**Author:** [@Marlon\_Henrique\_Teix](https://discuss.ai.google.dev/u/Marlon_Henrique_Teix)\ **Replies:** 1\ **Last updated:** [August 21, 2024, 1:28am UTC](https://discuss.ai.google.dev/t/how-to-handle-oom-at-tfx-transform-component/31611 "2024-08-21T01:28:50Z")
I have constructed a TFX pipeline but as I have increased the number of audio samples from 350 to 5000 I got OOM at the Transform component. I haven’t found parameters through which I can handle this problem. Here is my … --- ## [How to configure input as string paramenter with custom component?](https://discuss.ai.google.dev/t/how-to-configure-input-as-string-paramenter-with-custom-component/24061)
**Author:** [@Natielle\_Goncalves](https://discuss.ai.google.dev/u/Natielle_Goncalves)\ **Replies:** 1\ **Last updated:** [November 29, 2023, 2:30am UTC](https://discuss.ai.google.dev/t/how-to-configure-input-as-string-paramenter-with-custom-component/24061 "2023-11-29T02:30:13Z")
I updated the TFX from 0.22.0 to 1.8.0 and now I’m getting this error: TypeError: Argument input\_params should be a Channel of type \ (got test\_string) With 0.22.0 … --- ## [TFX pipeline root not found](https://discuss.ai.google.dev/t/tfx-pipeline-root-not-found/31720)
**Author:** [@Nazgyl77](https://discuss.ai.google.dev/u/Nazgyl77)\ **Replies:** 1\ **Last updated:** [November 29, 2023, 2:26am UTC](https://discuss.ai.google.dev/t/tfx-pipeline-root-not-found/31720 "2023-11-29T02:26:20Z")
Hello guys, im currently configuring a tfx pipeline with components such as SchemaGen,StatisticGen and Transform in my .ipynb: module\_file = os.path.abspath(“…\\components\\module.py”) #%% stats\_gen = StatisticsGen( e… --- ## [TFX Transform layer returning an empty dictionary](https://discuss.ai.google.dev/t/tfx-transform-layer-returning-an-empty-dictionary/31939)
**Author:** [@Marlon\_Henrique\_Teix](https://discuss.ai.google.dev/u/Marlon_Henrique_Teix)\ **Replies:** 5\ **Last updated:** [November 10, 2023, 10:31am UTC](https://discuss.ai.google.dev/t/tfx-transform-layer-returning-an-empty-dictionary/31939 "2023-11-10T10:31:18Z")
I have created a data pipeline. I have run all the components until my Trainer, in which I have an error in the function serve\_tf\_examples\_fn. More precisely, when transforming the raw data using the transform graph, it … --- ## [How to load a saved Example\_Gen artifact properly?](https://discuss.ai.google.dev/t/how-to-load-a-saved-example-gen-artifact-properly/13671)
**Author:** [@Marlon\_Henrique\_Teix](https://discuss.ai.google.dev/u/Marlon_Henrique_Teix)\ **Replies:** 0\ **Last updated:** [August 30, 2023, 2:13pm UTC](https://discuss.ai.google.dev/t/how-to-load-a-saved-example-gen-artifact-properly/13671 "2023-08-30T14:13:19Z")
I have constructed a pipeline using TFX for audio processing. It starts with a custom ExampleGen for audio processing and goes on the common way - statistics\_gen, Transform, and Trainer. Everything goes fine but increasi… --- ## [TFX Issue: RuntimeError: Failed to apply: CreateSavedModel\[tf\_v2\_only\]](https://discuss.ai.google.dev/t/tfx-issue-runtimeerror-failed-to-apply-createsavedmodel-tf-v2-only/24472)
**Author:** [@HIMANSHU\_SHAKYAWAR](https://discuss.ai.google.dev/u/HIMANSHU_SHAKYAWAR)\ **Replies:** 4\ **Last updated:** [August 16, 2023, 9:28pm UTC](https://discuss.ai.google.dev/t/tfx-issue-runtimeerror-failed-to-apply-createsavedmodel-tf-v2-only/24472 "2023-08-16T21:28:38Z")
Hello all, I am Trying to Train a Text Regression Model Using TFX Pipeline I have Done Multiple Experiments wrt to preprocessing\_fn but unfortunately Pipeline Breaks after Transform Component. Can anyone help me out. d… --- ## [Having NoneType error in multiple pipeline build using TFX and Airflow](https://discuss.ai.google.dev/t/having-nonetype-error-in-multiple-pipeline-build-using-tfx-and-airflow/31908)
**Author:** [@Bhargav\_Patel](https://discuss.ai.google.dev/u/Bhargav_Patel)\ **Replies:** 1\ **Last updated:** [January 24, 2023, 10:51pm UTC](https://discuss.ai.google.dev/t/having-nonetype-error-in-multiple-pipeline-build-using-tfx-and-airflow/31908 "2023-01-24T22:51:21Z")
I am working with TFX and AIrflow to build the pipeline for image classification and segmentation. I am replicating the repo by Jan Marcel Kezmann (Link). While executing the classification\_dag.py in dags/classification\_… --- ## [TFX integration with sparknlp models](https://discuss.ai.google.dev/t/tfx-integration-with-sparknlp-models/25388)
**Author:** [@reuth\_vexler](https://discuss.ai.google.dev/u/reuth_vexler)\ **Replies:** 0\ **Last updated:** [August 9, 2022, 2:18am UTC](https://discuss.ai.google.dev/t/tfx-integration-with-sparknlp-models/25388 "2022-08-09T02:18:20Z")
Hello, I’m interested in integrating sparknlp training pipelines into TFX Trainer, is this possible? is there any good explanation/blog for how to do this integration if possible? --- ## [Segmentation Fault in TensorFlow Transform component](https://discuss.ai.google.dev/t/segmentation-fault-in-tensorflow-transform-component/31317)
**Author:** [@gorjant](https://discuss.ai.google.dev/u/gorjant)\ **Replies:** 1\ **Last updated:** [July 31, 2022, 10:20am UTC](https://discuss.ai.google.dev/t/segmentation-fault-in-tensorflow-transform-component/31317 "2022-07-31T10:20:20Z")
Hi! I am trying the Taxi Tutorial on TFX 1.9 but getting a Segmentation Fault in the TensorFlow Transform Component (TFT). I have pinpointed the segfault happening during the execution of… tft.compute\_and\_apply\_vocabula… --- ## [How to add multiple pre-processing steps and a post-processing step for text-classifiction model to serve via tensorflow-serving?](https://discuss.ai.google.dev/t/how-to-add-multiple-pre-processing-steps-and-a-post-processing-step-for-text-classifiction-model-to-serve-via-tensorflow-serving/32052)
**Author:** [@naveen\_marthala](https://discuss.ai.google.dev/u/naveen_marthala)\ **Replies:** 5\ **Last updated:** [December 28, 2021, 5:32am UTC](https://discuss.ai.google.dev/t/how-to-add-multiple-pre-processing-steps-and-a-post-processing-step-for-text-classifiction-model-to-serve-via-tensorflow-serving/32052 "2021-12-28T05:32:25Z")
What I currently have and trying to do: When I receive a request from a client to the model in tensorflow-serving, I first need to process them using 13 regexes, then pass that text through tf.keras.preprocessing.text.T… --- ## [Does the CRFModelWrapper in the TFA support the serialization for the continuous training?](https://discuss.ai.google.dev/t/does-the-crfmodelwrapper-in-the-tfa-support-the-serialization-for-the-continuous-training/31291)
**Author:** [@dada\_Lai](https://discuss.ai.google.dev/u/dada_Lai)\ **Replies:** 9\ **Last updated:** [December 4, 2021, 9:13pm UTC](https://discuss.ai.google.dev/t/does-the-crfmodelwrapper-in-the-tfa-support-the-serialization-for-the-continuous-training/31291 "2021-12-04T21:13:49Z")
Hi everyone, I am using the CRFModelWrapper method following the tutorial as addons/layers\_crf.ipynb at add\_crf\_tutorial · howl-anderson/addons · GitHub to implement a Bi-LSTM -CRF neural-network for a multi-classes NER… --- ## [TFX ML Metadata Client project proposal](https://discuss.ai.google.dev/t/tfx-ml-metadata-client-project-proposal/23972)
**Author:** [@gcasassaez](https://discuss.ai.google.dev/u/gcasassaez)\ **Replies:** 0\ **Last updated:** [May 19, 2021, 4:40pm UTC](https://discuss.ai.google.dev/t/tfx-ml-metadata-client-project-proposal/23972 "2021-05-19T16:40:20Z")
Adding here a discussion around TFX ML Metadata Client proposal. TFX MLMD Client Library proposal by casassg · Pull Request #23 · tensorflow/tfx-addons · GitHub Currently under review in the PR above. --- ## [Help build new things for TFX!](https://discuss.ai.google.dev/t/help-build-new-things-for-tfx/25798)
**Author:** [@robertcrowe](https://discuss.ai.google.dev/u/robertcrowe)\ **Replies:** 0\ **Last updated:** [May 6, 2021, 6:26pm UTC](https://discuss.ai.google.dev/t/help-build-new-things-for-tfx/25798 "2021-05-06T18:26:10Z")
Do you have any ideas for new components, examples, or tools that can be added to TFX and shared with the world? If so, then you’re in luck! You can add your idea as a TFX-Addons community project idea by creating an iss…