Use parameters + buffers when deciding upscale_dtype#9882
Use parameters + buffers when deciding upscale_dtype#9882yiyixuxu merged 2 commits intohuggingface:mainfrom
Conversation
Sometimes, the decoder might lack parameters and only buffers (e.g., this happens when we manually need to convert all the parameters to buffers — e.g. to avoid packing fp16 and fp32 parameters with FSDP)
|
Could you maybe elaborate a bit more where this is useful? |
|
@universome a gentle ping |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
@universome Would it be possible to handle the merge conflict? We can merge this after. |
|
Hi @DN6 , sorry for the late reply. Basically, my use-case is the following. |
Sometimes, the decoder might lack parameters and only buffers (e.g., this happens when we manually need to convert all the parameters to buffers — e.g. to avoid packing fp16 and fp32 parameters with FSDP)
Sometimes, the decoder might lack parameters and only buffers (e.g., this happens when we manually need to convert all the parameters to buffers — e.g. to avoid packing fp16 and fp32 parameters with FSDP)
Sometimes, the decoder might lack parameters and only buffers (e.g., this happens when we manually need to convert all the parameters to buffers — e.g. to avoid packing fp16 and fp32 parameters with FSDP)
What does this PR do?
Fixes # (issue)
Before submitting
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.