new_qtensor support privateuseone allocator. #111464
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/111464
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 78338f2 with merge base 93a9b13 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
jbschlosser
added
the
triaged
|
@supriyar Could you take a look at this pr? |
|
can you add relevant testing to accompany this code change? cc @jerryzh168 to help review |
I've added a unittest for this case. Because quantize's calculations are complex and are registered elsewhere, I've just written an empty function for |
|
@supriyar @jerryzh168 please take a look at this pr. |
|
@huihoaan what is use case for this? we are actually moving away from having native quantized tensor in pytorch core going forward |
Our customer has a strong will to use this interface，the first step is to allow us to create a quantized tenser. So |
|
when it's something innocuous like this I just accept it |
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
I want to create a quant tensor through `PerTensorAffineQuantizer`. But I found that it will throw error because of the lake of judgment for PrivateUse1. Pull Request resolved: pytorch#111464 Approved by: https://github.com/ezyang
I want to create a quant tensor through `PerTensorAffineQuantizer`. But I found that it will throw error because of the lake of judgment for PrivateUse1. Pull Request resolved: pytorch#111464 Approved by: https://github.com/ezyang
7 participants
I want to create a quant tensor through
PerTensorAffineQuantizer. But I found that it will throw error because of the lake of judgment for PrivateUse1.