Optimize synthetic_data for large datasets#99
Merged
ryan112358 merged 2 commits intomasterfrom Feb 13, 2026
Merged
Conversation
Refactored the `synthetic_data` method in `MarkovRandomField` to significantly reduce memory usage. Previously, the method broadcasted conditional CDFs to all rows based on their parent configuration, resulting in O(N * D) memory usage where N is the number of rows and D is the domain size of the attribute being generated. This caused OOM errors for large N and large D. The new implementation: 1. Identifies unique parent configurations using `np.unique`. 2. Computes conditional CDFs only for these unique configurations. 3. Groups rows by parent configuration. 4. Uses `np.searchsorted` to perform inverse transform sampling for each group, avoiding the need to materialize the full N x D array. This approach scales roughly as O(U * D + N), where U is the number of unique parent configurations, which is typically much smaller than N. Verified with existing tests and a reproduction script based on issue #98. Co-authored-by: ryan112358 <8495634+ryan112358@users.noreply.github.com>
Contributor
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with New to Jules? Learn more at jules.google/docs. For security, I will only act on instructions from the user who triggered this task. |
Refactored the `synthetic_data` method in `MarkovRandomField` to significantly reduce memory usage. Previously, the method broadcasted conditional CDFs to all rows based on their parent configuration, resulting in O(N * D) memory usage where N is the number of rows and D is the domain size of the attribute being generated. This caused OOM errors for large N and large D. The new implementation: 1. Identifies unique parent configurations using `np.unique`. 2. Computes conditional CDFs only for these unique configurations. 3. Groups rows by parent configuration. 4. Uses `np.searchsorted` to perform inverse transform sampling for each group, avoiding the need to materialize the full N x D array. This approach scales roughly as O(U * D + N), where U is the number of unique parent configurations, which is typically much smaller than N. Benchmark Findings: - JIT Compilation (N=1): ~27.5s (vs 11.6s baseline) - N=1000: 0.27s (vs 0.17s) - N=10000: 1.09s (vs 0.90s) - N=100000: 7.46s (vs 8.58s) - N=1M: Runs without OOM, though possibly slower due to overhead for small domains. The memory reduction is the primary benefit, enabling scalability to large N and D where the previous implementation would fail. Co-authored-by: ryan112358 <8495634+ryan112358@users.noreply.github.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.
Addressed memory scalability issue in
MarkovRandomField.synthetic_databy replacing the dense broadcasting approach with a batched, unique-parent-based approach usingnp.searchsorted. This prevents OOM errors when generating large datasets with large-cardinality attributes.PR created automatically by Jules for task 8841611121634811676 started by @ryan112358