Comments (5)
From @kevinwuTT :
- On a completely fresh python virtual environment, I can successfully install ttnn from the github direct whl link and other requirements from pytorch2.0_ttnn requirements.txt. I can import ttnn successfully from here.
- However, the tests do depend on tests.ttnn.utils_for_testing module which isn't in the release whl from github. I'm not sure if I can remove this dependency since many of the unit tests from tt-metal/ttnn also uses the functions from that module.
- I will include a dev version of requirements.txt which won't install the whl and rely on the user creating the python virtual env with ./create_venv.sh from tt-metal. That creates an egg link to the ttnn source directory which makes development easier.
- I can push the changes to requirements.txt to main, but I am still not sure how to address the issue with the tests module from the first point.
from tt-metal.
tests.ttnn.utils_for_testing
is currently only used for a pcc
check.
We agreed that @kevinwuTT will remove it.
from tt-metal.
![Screenshot 2024-06-19 at 8 11 21 PM](https://private-user-images.githubusercontent.com/169092593/341226926-d8a2227b-aa9c-4ca7-a971-a46034fabfe3.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MjE2MjgzNTUsIm5iZiI6MTcyMTYyODA1NSwicGF0aCI6Ii8xNjkwOTI1OTMvMzQxMjI2OTI2LWQ4YTIyMjdiLWFhOWMtNGNhNy1hOTcxLWE0NjAzNGZhYmZlMy5wbmc_WC1BbXotQWxnb3JpdGhtPUFXUzQtSE1BQy1TSEEyNTYmWC1BbXotQ3JlZGVudGlhbD1BS0lBVkNPRFlMU0E1M1BRSzRaQSUyRjIwMjQwNzIyJTJGdXMtZWFzdC0xJTJGczMlMkZhd3M0X3JlcXVlc3QmWC1BbXotRGF0ZT0yMDI0MDcyMlQwNjAwNTVaJlgtQW16LUV4cGlyZXM9MzAwJlgtQW16LVNpZ25hdHVyZT0zOTU4ZjM5ZWEzMzkxZDAzMzNmMTdhOWU1MzEwYmZlNTcyMzJkODZmMzJlZGJiOWU1MmI5YjQ3MTFkZTE0YzU5JlgtQW16LVNpZ25lZEhlYWRlcnM9aG9zdCZhY3Rvcl9pZD0wJmtleV9pZD0wJnJlcG9faWQ9MCJ9.h4BtX8lWbTiLUwtCAnU9HqJpk2U3Q1dYt1D3lOq7M2w)
from tt-metal.
tt-metal-pytorch-vm-1 CI VM added to repo with sample GH workflow provided
tenstorrent/pytorch2.0_ttnn#4
https://github.com/tenstorrent-metal/pytorch2.0_ttnn/actions
Will need follow-up from @ayerofieiev-tt for list of deps needed for installation.
from tt-metal.
@tt-rkim
@ayerofieiev-tt has finished shifting the pytorch repo to the TT org and current members are ported over to have same access.
Would be prudent to start planning VM consolidation soon
from tt-metal.
Related Issues (20)
- Reduce duplication between Documentation publishing to central site and regular documentation for every commit to main
- Explore better way of building external docs
- Klassify model unit test
- Track and report number of retrains for each Eth Link during a test HOT 3
- Port over missing kernel variants from tt_eager binary to ttnn binary
- [Blackhole Bringup] Typecast failures
- New Op: All Gather Matmul HOT 2
- [llama] Add llama3 perplexity test to T3K perplexity pipeline
- Move tt_eager folder to ttnn/experimental
- Ubuntu 22.04 upgrade HOT 9
- [Bug Report] Cannot compile the latest `main` branch on GCC (control flow and designated initializer errors)
- [Feature Request] Support ttnn `moreh_layernorm`
- Migrate Where op
- Add sweeps and API calls for ttnn_unary_le_bw
- Add more explicit notes about isolated Python environments in INSTALLING.md so users aren't confused if they have a polluted system Python environment HOT 1
- Watcher kernel reporting doesn't say erisc
- RTArgs don't appear to work for slow dispatch on idle ethernet cores
- Improve e2e perf for Flash Decode
- [Blackhole Bringup] Test Unpack_tilize_A_B
- CI: build with address sanitization
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 tt-metal.