serving

serving copied to clipboard

We are following the document to build an optimized tf-serving docker image. Following the Dockerfile.devel and at the step of: bazel build --color=yes --curses=yes --verbose_failures --output_filter=DONT_MATCH_ANYTHING --config=release tensorflow_serving/model_servers:tensorflow_model_server there is...

fishfl

Fix a subset of existing lint issues

3

This PR fixes a bunch of existing lint issues that were automatically fixed by `ruff --fix`, in preparation for merging #2251. I've ignored a bunch of rule violations, but marked...

peytondmurray

When I run TF Serving on an x64 machine, I notice that TensorFlow uses brgemm_matmul_t for inference, while on an ARM architecture machine, it uses arm_gemm. How can I also...

nanzh-19

## URL(s) with the issue: Please provide a link to the documentation entry: https://tensorflow.github.io/serving/ ## Description of issue (what needs changing): After merging PR #2254, there are still bona fide...

smokestacklightnin

This PR adds a workflow to run pre-commit on every PR; a sensible initial set of pre-commit hooks is also included, and a new optional dependency has been added for...

peytondmurray

I use latest docker mirror tensorflow/serving:2.19.0-gpu on the Nvidia RTX 5090D The following error was reported: W external/org_tensorflow/tensorflow/core/framework/op_kernel.cc:1829] INTERNAL: 'cuLaunchKernel(function, gridX, gridY, gridZ, blockX, blockY, blockZ, 0, reinterpret_cast(stream), params, nullptr)'...

reindeergao

Question: How can I log timestamps for request lifecycle events (received, inference started, response sent)?

1

Hi, I'm new to this project and trying to better understand how to track the timing of inference requests inside the container. Specifically, I’d like to add logging or timestamps...

AnasHamrouni

Docker Image Operating System EoL on May 31, 2025

4

Hello! I noticed when I pulled the latest image from Docker Hub (`docker pull tensorflow/serving:latest`) that the image is using Ubuntu 20.04. The end of standard support for Ubuntu 20.04...

wildcardops

Update Dockerfiles to Ubuntu 22.04

1

The end of support for Ubuntu 20.04 is May 31st 2025. Updating the version of Ubuntu in the Dockerfiles to Ubuntu 22.04.

wildcardops

Error building TF Serving from source

4

Hi, I used the build command in this Installation [guide](https://www.tensorflow.org/tfx/serving/setup) `tools/run_in_docker.sh bazel build -c opt tensorflow_serving/...` But got this error:  Could you please help fix? Thanks!

nhatleSummer22