[Web] Implement linear congruential generator, make runtime seedable #16722
Conversation
This file contains 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
MasterJH5574
approved these changes
Mar 15, 2024
CharlieFRuan
added a commit
to mlc-ai/web-llm
that referenced
this pull request
Mar 15, 2024
This version supports seeding generation, for more see: - #335 On TVMjs side, we make the runtime seedable via: - apache/tvm#16722
thaisacs
pushed a commit
to thaisacs/tvm
that referenced
this pull request
Apr 3, 2024
…pache#16722) This PR implements `LinearCongruentialGenerator` in TVMjs, following the C++ counterpart in apache#8642. The motivation is that we want to seed autoregressive generation to make results reproducible, supporting the OpenAI field `seed`. The main function is `nextInt()`, which generates a number `(0, 2^32 - 1)` non-inclusive. Subsequently, we change all `Math.random()` in `runtime.ts` to `this.rng.randomFloat()`, exposing API `Instance.setSeed()`. Unit tests are added for `LinearCongruentialGenerator` for testing seed and coverage.
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Outdated suggestions cannot be applied.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 PR implements
LinearCongruentialGeneratorin TVMjs, following the C++ counterpart in #8642. The motivation is that we want to seed autoregressive generation to make results reproducible, supporting the OpenAI fieldseed. The main function isnextInt(), which generates a number(0, 2^32 - 1)non-inclusive.Subsequently, we change all
Math.random()inruntime.tstothis.rng.randomFloat(), exposing APIInstance.setSeed().Unit tests are added for
LinearCongruentialGeneratorfor testing seed and coverage.