docs: plan the multi_galaxy/group/cluster lensing-regime split (epic + 9 child prompts) - #104
Merged
Merged
Conversation
…+ 9 child prompts) Re-plan the too-large split_lensing_regimes intake into an executable epic: - Rename the parent prompt (multi_galaxy_lens -> multi_galaxy, mirroring the planned autogalaxy package name) and rewrite it as the coordinating plan: regime ladder taxonomy (all groups/clusters are multi-galaxy systems, not vice versa), literature research grounding each regime (verified arXiv IDs; flagships: SDSS J1011+0143 for multi_galaxy, CSWA 19 for the group halo-choice tutorial, Abell 2744/Bergamini 2023 confirmed for cluster), current-state audit, the expert-review conventions checklist, and the child-prompt execution order. - File 9 PR-sized child prompts: autolens_workspace multi_galaxy package, group halo-as-explicit-choice rework, cluster point-source-default narrative + extended-source follow-up, PyAutoLens RTD three-regime restructure, galaxy-scale extra/scaling feature parity, autogalaxy multi_galaxy + cluster packages, PyAutoGalaxy RTD guide updates, and the deferred assistants extension. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR
…llow-up The cluster-modeling expert corrections (vanishing unscaled member cores, Bergamini+19 tied exponents with gamma=0.2, r_cut_ref ~5", sigma_LT/sigma_0 attribution per the H. Ding derivation note) are now applied to the two mass-parameterization guides and the dPIE library docstrings on this branch. Update the epic plan checklist accordingly, sharpen the cluster child prompt with the remaining dataset-coupled sweep (start_here/simulator/modeling + regenerated cluster/simple dataset), and file a small follow-up feature prompt for an optional sigma_0-parameterized dPIE per the note's recommendation. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR
…branch Workspace package (start_here/simulator/modeling/READMEs, smoke-validated), workspace_test mirror (model_fit + relocated composition test), and the New User Guide four-rung ladder all landed 2026-07-25; prompt now tracks the remaining work (features scripts, real J1011+0143 data swap-in, likelihood_function/fit mirrors). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR
…culty key, track jax_likelihood variant Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR
Jammy2211
pushed a commit
that referenced
this pull request
Aug 10, 2026
Phase 2 is issued as autolens_profiling#103 with the schema leg in PR #104. Registered as PARTIAL rather than in-flight: pins and dashboard rows remain, and the Brain leg is blocked on PyAutoBrain#219 merging. Also records a CI gap found while working: autolens_profiling runs no pytest step, so scripts/misc/test/ never runs in CI and a vram_config test is failing on main unnoticed. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01L4STU81pQP1GkMzZVvsMsv
This was referenced Aug 10, 2026
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
2 participants
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.
Re-plan the too-large split_lensing_regimes intake into an executable epic:
planned autogalaxy package name) and rewrite it as the coordinating plan:
regime ladder taxonomy (all groups/clusters are multi-galaxy systems, not
vice versa), literature research grounding each regime (verified arXiv IDs;
flagships: SDSS J1011+0143 for multi_galaxy, CSWA 19 for the group
halo-choice tutorial, Abell 2744/Bergamini 2023 confirmed for cluster),
current-state audit, the expert-review conventions checklist, and the
child-prompt execution order.
group halo-as-explicit-choice rework, cluster point-source-default
narrative + extended-source follow-up, PyAutoLens RTD three-regime
restructure, galaxy-scale extra/scaling feature parity, autogalaxy
multi_galaxy + cluster packages, PyAutoGalaxy RTD guide updates, and the
deferred assistants extension.
Co-Authored-By: Claude Fable 5 noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_012Eacj9naBWSKrevsqnVJWR