Use the file size while reading onnx models. Ensure models are loaded using APIs in model.h for consistency.#4399
Merged
pranavsharma merged 4 commits intomasterfrom Jul 3, 2020
Merged
Conversation
… using APIs in model.h for consistency.
Contributor
Author
|
This still leaves the Load function that takes model_istream unoptimized as I'm not sure if there's a reliable way to get the size of the file given an istream. We could potentially make the block size configurable, but that requires a whole gamut of changes that touches all bindings, C APIs, etc and I'm not sure if we really require it. |
pranavsharma
commented
Jul 1, 2020
| #if GOOGLE_PROTOBUF_VERSION >= 3002000 | ||
| FileInputStream input(fd); | ||
| size_t file_size = 0; | ||
| int block_size = -1; |
Contributor
Author
There was a problem hiding this comment.
snnn
reviewed
Jul 1, 2020
snnn
reviewed
Jul 1, 2020
snnn
previously approved these changes
Jul 2, 2020
snnn
approved these changes
Jul 2, 2020
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description: Use the file size while reading onnx models. Ensure models are loaded using APIs in model.h for consistency.
Motivation and Context
This serves as a middle ground between providing a large fixed sized buffer of 4mb (done here and here) and the default of 8kb inside protobuf. The 4mb change causes an unnecessary increase in memory usage for even the smallest of the models due to this and the default is apparently slowing down model loading on Azure machines due to it's slow reads.