add int16/uint16 in quantization for cntk quantized speech model. #163
Conversation
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
snnn
approved these changes
Dec 13, 2018
liwchang
approved these changes
Dec 13, 2018
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This is adding int16/uint16 in quantization, since CNTK quantized speech model is using 16-bit in the first FC layer to ensure the accuracy. The PR is also clarifying the overflow policy, 1) production MUST not overflow, 2) Accumulation may overflow in 32 bits if the input is 8 bits or in 64 bits if the input is 16 bits.
@KeDengMS @liwchang let me know your thoughts please, especially the overflow clarification. Should we simplify it as "Accumulation may overflow in 32 bits in any case."? Thanks!