Open
Conversation
nkundiushuti
requested review from
GaganNarula and
Copilot
and removed request for
david-rx
There was a problem hiding this comment.
Pull request overview
This PR updates universal_torch_load() to cache cloud-hosted checkpoints by default, improving repeated model load times (notably for hf:// paths) by persisting downloads under the ESP cache directory.
Changes:
- Change
universal_torch_load()defaultcache_modefrom"none"to"use". - Improve cloud-cache keying by hashing the full URI and organizing cached files under a per-
bucketsubdirectory to reduce filename collisions. - Update docstring to reflect the new default caching behavior.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
nkundiushuti
changed the title
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 makes universal_torch_load() default to cache_mode="use" so Hugging Face (hf://) checkpoints are saved to the local ESP_CACHE_HOME/~/.cache/esp cache and reused across runs.
Impact: second avex.load_model(...) call is much faster (measured ~5× warm-load speedup after a cold download).
Validation: pytest full suite passed (277 passed, 7 skipped) and all examples/*.py using load_model ran successfully on CPU.
fix for #171