docs: add training recommendations for epochs, shots, and noise upscaling#71
Closed
ivanbasov wants to merge 4 commits into
Closed
docs: add training recommendations for epochs, shots, and noise upscaling#71ivanbasov wants to merge 4 commits into
ivanbasov wants to merge 4 commits into
Conversation
…fault torch.compile=on combined with DataLoader spawn workers during LER validation causes a segfault (20 leaked semaphores, core dumped). Set PREDECODER_TORCH_COMPILE=0 for the Train all orientations step. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…vent segfault" This reverts commit 7f0f6c8.
…ling - Advise ≥100 epochs for models 1, 4, 5 (uncorrelated matching) in both README and TRAINING env-var table - Document 67 M shots/epoch with 8 GPUs as the recommended sample count - Add concise algorithm summary to the noise upscaling section (p_max → rescale to 0.6 %) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
README.mdmodel-selection section and theTRAINING.mdenv-var table (PREDECODER_TRAIN_EPOCHS).PREDECODER_TRAIN_SAMPLES=67108864); fewer shots produce worse results. Added in the same two locations.README.md, explicitly stating the p_max → rescale-to-0.6% recipe so the mechanism is clear at a glance.Test plan
spdx-header-checkunaffected (no source files changed).🤖 Generated with Claude Code