Comments (10)
So true! We want to set something up with Jenkins, like the rest of TensorFlow has. @gunan maybe you can help us get started?
from haskell.
Let me consult a few people here to see where we can run tests on this repo.
I will assign the issue myself to not forget.
from haskell.
I cannot seem to assign the issue :)
I will update here with more information.
from haskell.
OK, I can help you get some tests running on ci.tensorflow.org
You simply need to give me a bash script I need to run, Then I will setup the continuous build on master.
Then we need to setup some webhooks. We need a repository administrator to do that.
I can also do this wiring if you give me temporary access.
from haskell.
I imagine the easiest would be for us to have a canonical docker image which rebuilds everything including TF? Something akin to Mac OS build instructions in our README.md? Once I have this checked in we can ask @martinwicke to grant the necessary permissions to @gunan. I'm taking over this issue to do the first step.
from haskell.
If you only would like a build, it is OK.
However, if you actually have some tests that need to be run, that would be even better.
One other option is to:
- fetch the latest TF docker image,
- start a container with this image
- inside that container run a script that builds and tests haskell bindings.
I think this would be even faster to run as we avoid building full TF.
from haskell.
We do have the tests. But I don't know if there's an option to communicate individual test status beyond a simple exit code from the script. Probably not really what you had in mind @gunan? One thing about using a "raw" TF container is we have a pretty heavy build dependencies that would be nice to cache (the compiler and all the dependent libraries).
from haskell.
I see, then how about building a new Docker image on top of TF docker?
Like how we do on top of nvidia/cuda
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/docker/Dockerfile.gpu#L1
I am just trying to get you a faster test by avoiding rebuild of TF.
For the test script, are these bazel tests?
We have some bash scripts to run and capture all test outputs, maybe we can prepare something like that for this repo too?
from haskell.
They are not bazel tests (pragmatic reasons). Pattern matching should be easy, but let's cross that bridge later.
from haskell.
#14 has the script: ./ci_build/outer_launch_tests.sh
from haskell.
Related Issues (20)
- Reverse-mode automatic differentiation HOT 1
- 404 on instructions Stack + Docker + GPU HOT 1
- SIGSEGV when computing gradients for reduceMean using placeholders
- tensorflow-core-ops won't compile HOT 6
- Why is tensorflow-mnist-input-data not on Hackage? HOT 5
- tensorflow-ops compile error HOT 1
- Fail on Debian Build HOT 2
- Stack install - Symbolic link dest not found... HOT 6
- Package won't install. HOT 2
- gpu container cannot be created HOT 1
- tensorflow-core-ops compilation error HOT 2
- tensorflow-core-ops doesn't compile with tensorflor-2.4 HOT 3
- TensorData decoding problem with Nix package
- Re-add blackgnezdo HOT 1
- Library won't compile. HOT 2
- Tensorflow 2.9 Compatibility HOT 2
- Running a loaded TF2 SavedModel gives TensorfFlowException TF_INVALID_ARGUMENT HOT 2
- Nvidia (3000 series) Ampere Architecture Support - Update TensorFlow to v2.4.0 HOT 2
- Any plans to update the Hackage?
- problems getting started... HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from haskell.