labloop 0.1.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (39) hide show
  1. labloop-0.1.0/.gitignore +19 -0
  2. labloop-0.1.0/CHANGELOG.md +96 -0
  3. labloop-0.1.0/CONTRIBUTING.md +64 -0
  4. labloop-0.1.0/LICENSE +202 -0
  5. labloop-0.1.0/PKG-INFO +411 -0
  6. labloop-0.1.0/README.md +379 -0
  7. labloop-0.1.0/ROADMAP.md +142 -0
  8. labloop-0.1.0/experiments/outcome_granularity/RESULTS.md +78 -0
  9. labloop-0.1.0/experiments/outcome_granularity/raw_results.json +6523 -0
  10. labloop-0.1.0/experiments/outcome_granularity/simulate.py +354 -0
  11. labloop-0.1.0/pyproject.toml +79 -0
  12. labloop-0.1.0/src/labloop/__init__.py +55 -0
  13. labloop-0.1.0/src/labloop/brief.py +176 -0
  14. labloop-0.1.0/src/labloop/cli.py +574 -0
  15. labloop-0.1.0/src/labloop/integrity.py +173 -0
  16. labloop-0.1.0/src/labloop/ledger.py +158 -0
  17. labloop-0.1.0/src/labloop/lock.py +120 -0
  18. labloop-0.1.0/src/labloop/loop.py +592 -0
  19. labloop-0.1.0/src/labloop/metrics.py +95 -0
  20. labloop-0.1.0/src/labloop/py.typed +0 -0
  21. labloop-0.1.0/src/labloop/runner.py +92 -0
  22. labloop-0.1.0/src/labloop/types.py +190 -0
  23. labloop-0.1.0/src/labloop/workspace.py +145 -0
  24. labloop-0.1.0/tests/__init__.py +0 -0
  25. labloop-0.1.0/tests/conftest.py +97 -0
  26. labloop-0.1.0/tests/test_brief.py +195 -0
  27. labloop-0.1.0/tests/test_cli.py +371 -0
  28. labloop-0.1.0/tests/test_directions.py +285 -0
  29. labloop-0.1.0/tests/test_harness.py +123 -0
  30. labloop-0.1.0/tests/test_init_and_query.py +162 -0
  31. labloop-0.1.0/tests/test_integrity.py +137 -0
  32. labloop-0.1.0/tests/test_ledger.py +168 -0
  33. labloop-0.1.0/tests/test_lock.py +182 -0
  34. labloop-0.1.0/tests/test_loop.py +450 -0
  35. labloop-0.1.0/tests/test_metrics.py +77 -0
  36. labloop-0.1.0/tests/test_noise.py +142 -0
  37. labloop-0.1.0/tests/test_resume.py +186 -0
  38. labloop-0.1.0/tests/test_runner.py +118 -0
  39. labloop-0.1.0/tests/test_workspace.py +155 -0
@@ -0,0 +1,19 @@
1
+ __pycache__/
2
+ *.py[cod]
3
+ .venv/
4
+ venv/
5
+ dist/
6
+ build/
7
+ *.egg-info/
8
+ .pytest_cache/
9
+ .ruff_cache/
10
+ .coverage
11
+ htmlcov/
12
+
13
+ # labloop's own run output
14
+ labloop.jsonl
15
+
16
+ .DS_Store
17
+ .idea/
18
+ .vscode/
19
+ dist/
@@ -0,0 +1,96 @@
1
+ # Changelog
2
+
3
+ ## 0.1.0 — unreleased
4
+
5
+ First release. Everything below was built and then used in anger before
6
+ shipping: the loop was pointed at real experiment shapes — a regression
7
+ task, a timing benchmark, a checkpoint-heavy training script, a pure-noise
8
+ metric — and every bug that surfaced is listed here with the behavior that
9
+ replaced it.
10
+
11
+ ### The loop
12
+
13
+ - Keep-or-revert over any command and any metric: `propose` mutates the
14
+ tree, `run` prints a number, an improvement is committed and anything
15
+ else reverted. Ties revert — a tie is not an improvement.
16
+ - Nine trial outcomes, each a distinct event that sends you somewhere
17
+ different: `kept`, `reverted`, `no_change`, `failed`, `timed_out`,
18
+ `no_metric`, `not_finite`, `harness_changed`, `interrupted`. Measured
19
+ (`experiments/outcome_granularity/`): outcome labels beat kept/not-kept
20
+ on convergence with p < 0.01; the five-to-nine refinement is not
21
+ statistically distinguishable and the set stays for the bugs its
22
+ distinctions fixed.
23
+ - Separate proposal budget (`--propose-budget`): an agent thinking longer
24
+ than the experiment runs is ordinary. Timeouts kill the whole process
25
+ group either way.
26
+ - A run stops after ten consecutive trials with no verdict
27
+ (`--give-up-after`), instead of failing identically all night.
28
+
29
+ ### Trusting the number
30
+
31
+ - `labloop noise` measures whether the metric holds still before anything
32
+ is optimized; reports spread and standard deviation with the exact
33
+ follow-up command.
34
+ - `--min-delta` and `--confirm` for metrics that move on their own —
35
+ measured on pure noise over 400 replicated runs: fabricated improvement
36
+ falls from 23.5% to 9.9% with both, for 10% more compute. Neither makes
37
+ a noisy metric safe, and the docs say so.
38
+ - Non-finite metrics never become the incumbent: one `nan` used to freeze
39
+ the loop permanently, reverting every later improvement while `best`
40
+ reported nan.
41
+ - `--protect` digests the files that define the measurement (SHA-256,
42
+ per-file); a proposal that edits the evaluator or the held-out data is
43
+ recorded as `harness_changed` and named, not scored. The ledger itself
44
+ is checked on every trial without being declared. Detection, not
45
+ prevention — a claim the design can keep.
46
+ - An incumbent measured under a different harness digest stops the loop
47
+ instead of being compared against; metric or goal drift against an
48
+ existing ledger is refused with the field named.
49
+
50
+ ### The record
51
+
52
+ - Append-only JSONL ledger; half-written lines, unknown outcomes and
53
+ unknown fields are skipped, not fatal. Valid strict JSON always.
54
+ - A kept trial commits exactly the proposed change plus a sparse decision
55
+ log (`labloop-history.jsonl`, reverted trials included); run artifacts
56
+ are discarded, so a checkpoint-per-trial training run does not become a
57
+ repository of hundreds of gigabytes.
58
+ - Every run records its spec as a manifest line; `labloop resume`
59
+ continues under the last spec that can actually run — a baseline
60
+ re-measurement after the crash does not cost you the run spec. Same
61
+ incumbent, same numbering, nothing retyped. The environment is never
62
+ recorded.
63
+ - One loop per ledger, enforced by an OS-released lock; a second run is
64
+ refused with the holder's pid, `--wait` queues. A crashed run cannot
65
+ leave a stale lock.
66
+ - Research directions: fork from any kept trial (`labloop branch`), run
67
+ each in its own worktree against the shared ledger, per-direction
68
+ incumbents seeded from the fork point. `log --compare` refuses across
69
+ differing harnesses. A direction the ledger has never heard of is
70
+ refused with a did-you-mean — a typo'd `--direction` used to create a
71
+ phantom direction whose first trial, however bad, was kept.
72
+ - Querying without jq: `log --json`, `--outcome`, `--direction`,
73
+ `--since-trial`.
74
+
75
+ ### Feedback to the proposer
76
+
77
+ - Each proposal receives `$LABLOOP_BRIEF`: its direction's history with a
78
+ one-sentence `why` per trial, failures carrying their output tail;
79
+ scalars in `$LABLOOP_METRIC`, `$LABLOOP_GOAL`, `$LABLOOP_INCUMBENT`,
80
+ `$LABLOOP_TRIAL`. Written by labloop and read by the agent, never the
81
+ reverse.
82
+
83
+ ### Fixed before anyone else could hit them
84
+
85
+ Eleven bugs found by dogfooding, each now a regression test: the nan
86
+ freeze; a crash that printed nothing reported as `no_metric`; `nan` read
87
+ differently by the two output parsers; metric names ending in punctuation
88
+ missed in key=value form; a no-op proposal crashing the run on git's
89
+ empty-commit refusal; an interrupted trial vanishing from the ledger (and
90
+ recovery advice that would have committed an unmeasured change); a commit
91
+ refused by a pre-commit hook taking the run down with an empty ledger;
92
+ `revert` unable to undo staged changes (`git checkout -- .` restores from
93
+ the index), wedging the tree permanently; a relative `--ledger` resolving
94
+ against the shell's cwd and silently splitting the record; one budget
95
+ covering both the agent and the experiment; and a proposal killed at the
96
+ budget reported as failed rather than timed out.
@@ -0,0 +1,64 @@
1
+ # Contributing
2
+
3
+ ## Setup
4
+
5
+ ```bash
6
+ git clone https://github.com/foothills-labs/labloop
7
+ cd labloop
8
+ pip install -e ".[dev]"
9
+ ```
10
+
11
+ Stdlib only at runtime — the package has zero dependencies, and keeping it
12
+ that way is a feature. `pytest`, `ruff` and `mypy` are dev-only.
13
+
14
+ ## The checks
15
+
16
+ Everything CI runs, runnable locally:
17
+
18
+ ```bash
19
+ pytest -q # the suite; ~230 tests, under 30 seconds
20
+ ruff check . # lint
21
+ mypy # strict, src only; the package ships py.typed
22
+ ```
23
+
24
+ ## How changes are made here
25
+
26
+ **Tests come first.** Write the failing test, watch it fail, then write the
27
+ code that makes it pass. A test written after the fact tends to enshrine
28
+ whatever the code happens to do; a test written first states what it should
29
+ do. This repository's history has a worked example: a failing test for a
30
+ typo'd `--direction` flushed out four existing tests that were quietly
31
+ pinning behavior nobody wanted.
32
+
33
+ **Dogfood before you trust.** Most of the bugs fixed here were found by
34
+ using the tool on realistic tasks, not by reading the code. If you add a
35
+ user-facing behavior, drive it from the command line the way a stranger
36
+ would — including the mistakes a stranger would make.
37
+
38
+ **Measure claims.** House style is numbers over adjectives. If a change is
39
+ justified by convergence, noise robustness, or performance, the evidence
40
+ belongs in `experiments/` with statistics, not in the commit message as an
41
+ assertion. Negative results are published, not discarded — see
42
+ `experiments/outcome_granularity/RESULTS.md` for the shape.
43
+
44
+ **Settled design decisions.** Some behaviors look odd and are load-bearing.
45
+ Before "fixing" one, check whether a test asserts it on purpose:
46
+
47
+ - A tie is not an improvement; equal metrics revert.
48
+ - A missing metric is not a bad score; nine outcomes exist because each
49
+ sends the user somewhere different.
50
+ - The loop refuses to start on a dirty tree; it reverts by discarding.
51
+ - Kept commits contain the proposed change and the decision log, never
52
+ run artifacts.
53
+ - The ledger is append-only, valid strict JSON, and the source of truth
54
+ for the incumbent.
55
+ - The brief is written by labloop and read by the agent, never the
56
+ reverse.
57
+
58
+ Changing any of these is a design discussion, not a cleanup.
59
+
60
+ ## Commit messages
61
+
62
+ Explain *why*, in prose, the way the existing history does. The subject
63
+ line states what changed; the body states what was wrong, how it was found,
64
+ and what now holds. "Fixed bug" is not a commit message.
labloop-0.1.0/LICENSE ADDED
@@ -0,0 +1,202 @@
1
+
2
+ Apache License
3
+ Version 2.0, January 2004
4
+ http://www.apache.org/licenses/
5
+
6
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7
+
8
+ 1. Definitions.
9
+
10
+ "License" shall mean the terms and conditions for use, reproduction,
11
+ and distribution as defined by Sections 1 through 9 of this document.
12
+
13
+ "Licensor" shall mean the copyright owner or entity authorized by
14
+ the copyright owner that is granting the License.
15
+
16
+ "Legal Entity" shall mean the union of the acting entity and all
17
+ other entities that control, are controlled by, or are under common
18
+ control with that entity. For the purposes of this definition,
19
+ "control" means (i) the power, direct or indirect, to cause the
20
+ direction or management of such entity, whether by contract or
21
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
22
+ outstanding shares, or (iii) beneficial ownership of such entity.
23
+
24
+ "You" (or "Your") shall mean an individual or Legal Entity
25
+ exercising permissions granted by this License.
26
+
27
+ "Source" form shall mean the preferred form for making modifications,
28
+ including but not limited to software source code, documentation
29
+ source, and configuration files.
30
+
31
+ "Object" form shall mean any form resulting from mechanical
32
+ transformation or translation of a Source form, including but
33
+ not limited to compiled object code, generated documentation,
34
+ and conversions to other media types.
35
+
36
+ "Work" shall mean the work of authorship, whether in Source or
37
+ Object form, made available under the License, as indicated by a
38
+ copyright notice that is included in or attached to the work
39
+ (an example is provided in the Appendix below).
40
+
41
+ "Derivative Works" shall mean any work, whether in Source or Object
42
+ form, that is based on (or derived from) the Work and for which the
43
+ editorial revisions, annotations, elaborations, or other modifications
44
+ represent, as a whole, an original work of authorship. For the purposes
45
+ of this License, Derivative Works shall not include works that remain
46
+ separable from, or merely link (or bind by name) to the interfaces of,
47
+ the Work and Derivative Works thereof.
48
+
49
+ "Contribution" shall mean any work of authorship, including
50
+ the original version of the Work and any modifications or additions
51
+ to that Work or Derivative Works thereof, that is intentionally
52
+ submitted to Licensor for inclusion in the Work by the copyright owner
53
+ or by an individual or Legal Entity authorized to submit on behalf of
54
+ the copyright owner. For the purposes of this definition, "submitted"
55
+ means any form of electronic, verbal, or written communication sent
56
+ to the Licensor or its representatives, including but not limited to
57
+ communication on electronic mailing lists, source code control systems,
58
+ and issue tracking systems that are managed by, or on behalf of, the
59
+ Licensor for the purpose of discussing and improving the Work, but
60
+ excluding communication that is conspicuously marked or otherwise
61
+ designated in writing by the copyright owner as "Not a Contribution."
62
+
63
+ "Contributor" shall mean Licensor and any individual or Legal Entity
64
+ on behalf of whom a Contribution has been received by Licensor and
65
+ subsequently incorporated within the Work.
66
+
67
+ 2. Grant of Copyright License. Subject to the terms and conditions of
68
+ this License, each Contributor hereby grants to You a perpetual,
69
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70
+ copyright license to reproduce, prepare Derivative Works of,
71
+ publicly display, publicly perform, sublicense, and distribute the
72
+ Work and such Derivative Works in Source or Object form.
73
+
74
+ 3. Grant of Patent License. Subject to the terms and conditions of
75
+ this License, each Contributor hereby grants to You a perpetual,
76
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77
+ (except as stated in this section) patent license to make, have made,
78
+ use, offer to sell, sell, import, and otherwise transfer the Work,
79
+ where such license applies only to those patent claims licensable
80
+ by such Contributor that are necessarily infringed by their
81
+ Contribution(s) alone or by combination of their Contribution(s)
82
+ with the Work to which such Contribution(s) was submitted. If You
83
+ institute patent litigation against any entity (including a
84
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
85
+ or a Contribution incorporated within the Work constitutes direct
86
+ or contributory patent infringement, then any patent licenses
87
+ granted to You under this License for that Work shall terminate
88
+ as of the date such litigation is filed.
89
+
90
+ 4. Redistribution. You may reproduce and distribute copies of the
91
+ Work or Derivative Works thereof in any medium, with or without
92
+ modifications, and in Source or Object form, provided that You
93
+ meet the following conditions:
94
+
95
+ (a) You must give any other recipients of the Work or
96
+ Derivative Works a copy of this License; and
97
+
98
+ (b) You must cause any modified files to carry prominent notices
99
+ stating that You changed the files; and
100
+
101
+ (c) You must retain, in the Source form of any Derivative Works
102
+ that You distribute, all copyright, patent, trademark, and
103
+ attribution notices from the Source form of the Work,
104
+ excluding those notices that do not pertain to any part of
105
+ the Derivative Works; and
106
+
107
+ (d) If the Work includes a "NOTICE" text file as part of its
108
+ distribution, then any Derivative Works that You distribute must
109
+ include a readable copy of the attribution notices contained
110
+ within such NOTICE file, excluding those notices that do not
111
+ pertain to any part of the Derivative Works, in at least one
112
+ of the following places: within a NOTICE text file distributed
113
+ as part of the Derivative Works; within the Source form or
114
+ documentation, if provided along with the Derivative Works; or,
115
+ within a display generated by the Derivative Works, if and
116
+ wherever such third-party notices normally appear. The contents
117
+ of the NOTICE file are for informational purposes only and
118
+ do not modify the License. You may add Your own attribution
119
+ notices within Derivative Works that You distribute, alongside
120
+ or as an addendum to the NOTICE text from the Work, provided
121
+ that such additional attribution notices cannot be construed
122
+ as modifying the License.
123
+
124
+ You may add Your own copyright statement to Your modifications and
125
+ may provide additional or different license terms and conditions
126
+ for use, reproduction, or distribution of Your modifications, or
127
+ for any such Derivative Works as a whole, provided Your use,
128
+ reproduction, and distribution of the Work otherwise complies with
129
+ the conditions stated in this License.
130
+
131
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
132
+ any Contribution intentionally submitted for inclusion in the Work
133
+ by You to the Licensor shall be under the terms and conditions of
134
+ this License, without any additional terms or conditions.
135
+ Notwithstanding the above, nothing herein shall supersede or modify
136
+ the terms of any separate license agreement you may have executed
137
+ with Licensor regarding such Contributions.
138
+
139
+ 6. Trademarks. This License does not grant permission to use the trade
140
+ names, trademarks, service marks, or product names of the Licensor,
141
+ except as required for reasonable and customary use in describing the
142
+ origin of the Work and reproducing the content of the NOTICE file.
143
+
144
+ 7. Disclaimer of Warranty. Unless required by applicable law or
145
+ agreed to in writing, Licensor provides the Work (and each
146
+ Contributor provides its Contributions) on an "AS IS" BASIS,
147
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148
+ implied, including, without limitation, any warranties or conditions
149
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150
+ PARTICULAR PURPOSE. You are solely responsible for determining the
151
+ appropriateness of using or redistributing the Work and assume any
152
+ risks associated with Your exercise of permissions under this License.
153
+
154
+ 8. Limitation of Liability. In no event and under no legal theory,
155
+ whether in tort (including negligence), contract, or otherwise,
156
+ unless required by applicable law (such as deliberate and grossly
157
+ negligent acts) or agreed to in writing, shall any Contributor be
158
+ liable to You for damages, including any direct, indirect, special,
159
+ incidental, or consequential damages of any character arising as a
160
+ result of this License or out of the use or inability to use the
161
+ Work (including but not limited to damages for loss of goodwill,
162
+ work stoppage, computer failure or malfunction, or any and all
163
+ other commercial damages or losses), even if such Contributor
164
+ has been advised of the possibility of such damages.
165
+
166
+ 9. Accepting Warranty or Additional Liability. While redistributing
167
+ the Work or Derivative Works thereof, You may choose to offer,
168
+ and charge a fee for, acceptance of support, warranty, indemnity,
169
+ or other liability obligations and/or rights consistent with this
170
+ License. However, in accepting such obligations, You may act only
171
+ on Your own behalf and on Your sole responsibility, not on behalf
172
+ of any other Contributor, and only if You agree to indemnify,
173
+ defend, and hold each Contributor harmless for any liability
174
+ incurred by, or claims asserted against, such Contributor by reason
175
+ of your accepting any such warranty or additional liability.
176
+
177
+ END OF TERMS AND CONDITIONS
178
+
179
+ APPENDIX: How to apply the Apache License to your work.
180
+
181
+ To apply the Apache License to your work, attach the following
182
+ boilerplate notice, with the fields enclosed by brackets "[]"
183
+ replaced with your own identifying information. (Don't include
184
+ the brackets!) The text should be enclosed in the appropriate
185
+ comment syntax for the file format. We also recommend that a
186
+ file or class name and description of purpose be included on the
187
+ same "printed page" as the copyright notice for easier
188
+ identification within third-party archives.
189
+
190
+ Copyright 2026 Foothills Labs
191
+
192
+ Licensed under the Apache License, Version 2.0 (the "License");
193
+ you may not use this file except in compliance with the License.
194
+ You may obtain a copy of the License at
195
+
196
+ http://www.apache.org/licenses/LICENSE-2.0
197
+
198
+ Unless required by applicable law or agreed to in writing, software
199
+ distributed under the License is distributed on an "AS IS" BASIS,
200
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201
+ See the License for the specific language governing permissions and
202
+ limitations under the License.