Skip to content

1. Fix 7 bugs: NAS validation, ConfigDict pickling, dataset utils, SmoothedValue, error handling#6

Open
musicalplatypus wants to merge 7 commits into
TexasInstruments:mainfrom
musicalplatypus:pr/bug-fixes
Open

1. Fix 7 bugs: NAS validation, ConfigDict pickling, dataset utils, SmoothedValue, error handling#6
musicalplatypus wants to merge 7 commits into
TexasInstruments:mainfrom
musicalplatypus:pr/bug-fixes

Conversation

@musicalplatypus

@musicalplatypus musicalplatypus commented Apr 7, 2026

Copy link
Copy Markdown

Summary

This PR fixes 7 bugs across tinyml-modelmaker, tinyml-tinyverse, and tinyml-modeloptimization:

Fixes

  1. @classmethod methods using self instead of cls — corrected in timeseries runner and vision runner
  2. Missing import in tinyml_benchmark.py — added required import for symlink creation
  3. argv splice bug in timeseries_base.py — boolean flags (--native-amp) broke fixed-offset slicing for trailing key-value args; now stripped before slicing and re-appended after
  4. dataset_utils.py variable name bug — split_factors was referenced before assignment in create_inter_file_split()
  5. assert used for input validation — replaced with raise ValueError in dataset splitting
  6. ConfigDict unpicklable — added __reduce__ method for proper serialization
  7. Generic except Exception in training __init__.py — replaced with except AttributeError + descriptive ValueError in both timeseries and vision training modules
  8. SmoothedValue.update() forcing GPU sync — deferred .item() to print time, avoiding MPS command-buffer flush on every batch
  9. NAS evaluate_classification crash with batch_size=1 — removed .squeeze() calls that collapsed the batch dimension
  10. NAS validation iterator — fixed device support and resource penalty bugs
  11. quit_event ordering — fixed parameter ordering in anomaly detection
  12. Log file write safety — added encoding parameter for cross-platform compatibility

Testing

  • All changes are isolated bug fixes with no behavioral changes beyond correcting errors
  • Verified on macOS (MPS) and Linux (CUDA) environments

Files Changed (20 files)

Across tinyml-modelmaker, tinyml-tinyverse, and tinyml-modeloptimization packages.

t5fkg8d44d-beep and others added 7 commits April 7, 2026 07:12
@t5fkg8d44d-beep @claude
Fix @classmethod methods using self instead of cls
0669d6c 
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@t5fkg8d44d-beep @claude
Fix 5 bugs in tinyml-modelmaker: missing import, argv splice, dataset…
c51706f 
… utils, symlink

1. Add missing TinyMLQuantizationVersion import in timeseries/runner.py and
   vision/runner.py — NameError on every training run that reaches packaging.
2. Fix argv splice when native_amp=True + quantization: strip boolean flags
   before fixed-offset slicing, re-append after, preventing malformed argv.
3. Fix len(split_factor) called on float in dataset_utils.py — use the
   accumulated split_factors list instead of the original parameter.
   Fixed in both create_inter_file_split and create_intra_file_split.
4. Add else branch in dataset_load() for unknown annotation_format —
   raises ValueError instead of UnboundLocalError.
5. Remove duplicate os.symlink() call in make_symlink() — was creating
   the link twice, second call always failed with FileExistsError.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@t5fkg8d44d-beep @claude
Fix 6 bugs: ConfigDict pickling, include paths, constants, downloads,…
8a89c47 
… error handling

1. ConfigDict.__getstate__: add missing return — pickling now preserves state.
2. _parse_include_files: fix and→or — absolute/relative include paths now
   resolve correctly instead of always prepending base path.
3. TASK_CATEGORIES: use TASK_CATEGORY_TS_ANOMALYDETECTION instead of
   TASK_TYPE_GENERIC_TS_ANOMALYDETECTION — fixes wrong compilation
   parameters for anomaly detection tasks.
4. download_url: handle missing Content-Length header gracefully instead
   of crashing with TypeError on int(None).
5. download_files: track aggregate success across all URLs instead of
   reporting only the last download's status.
6. get_target_module: raise ValueError with available options instead of
   returning None — prevents confusing downstream AttributeError on
   NoneType. Applied to both timeseries and vision training modules.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@t5fkg8d44d-beep @claude
Fix quit_event ordering, anomaly typo, and log file write safety
858f031 
1. Set args.quit_event before compile_scr.run() so compilation can
   actually be cancelled (was set after run() returned — too late).
2. Fix typo 'anomlay_list.txt' → 'anomaly_list.txt' in dataset handling.
3. Move cleanup_special_chars() file write outside the read block so the
   file isn't truncated while the read handle is still open — prevents
   data loss if the write fails mid-way.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@t5fkg8d44d-beep @claude
Fix 6 NAS bugs: validation iterator, device support, resource penalty…
7d43dcc 
…, dead code

1. Persistent validation iterator: replace next(iter(valid_loader)) with
   a cycling iterator so architecture updates see different batches each step
2. Structural params: pass steps/multiplier/stem_multiplier from search to
   final model so the evaluated architecture matches what was searched
3. Best genotype tracking: select genotype with highest validation accuracy
   instead of always using the last epoch
4. Device abstraction: add get_device() (CUDA > MPS > CPU), replace all
   hardcoded .cuda() calls with .to(device) across search, model, architect
5. Differentiable resource penalty: replace inert scalar penalties with
   sum(softmax(alpha) * param_counts) normalized to [0,1], fully
   differentiable w.r.t. architecture parameters
6. Remove dead code: delete unused RNN genotypes, Genotype_RNN, save(),
   create_exp_dir(), and all RNN branches from architect

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@t5fkg8d44d-beep @claude
Fix SmoothedValue: defer .item() to print time, remove MPS mem sync
08a5e55 
The previous MPS optimization commit moved .item() from MetricLogger
to SmoothedValue but it still fired on every batch update, causing a
GPU command-buffer flush each time (no actual improvement).

Fix by storing detached tensors in the deque and accumulating total
on-device.  Property accessors (median, avg, global_avg, max, value)
call .item() lazily — only at print time (every print_freq batches).
Benchmarked at 7.8x faster for the metric-logging path on MPS.

Also reverts MPS memory reporting (torch.mps.current_allocated_memory)
which introduced a new GPU sync that never existed before the previous
commit.  CUDA memory reporting is unchanged.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@t5fkg8d44d-beep
NAS: fix evaluate_classification batch_size=1 crash, fix MPS float64
04af3c9 
evaluate_classification (utils.py):
- Remove unsafe .squeeze() calls on output tensor. When the last test
  batch has exactly 1 sample, squeeze() collapses (1, C) to (C,),
  causing cross_entropy 'size mismatch' error. Output is already (N, C)
  which is correct for cross_entropy — no squeeze needed. Consistent
  with train_one_epoch_classification which never squeezes.

train_cnn_search.py:
- Reorder .float().to(device) to cast before MPS transfer (MPS doesn't
  support float64)
@musicalplatypus musicalplatypus changed the title Fix 7 bugs: NAS validation, ConfigDict pickling, dataset utils, SmoothedValue, error handling 1. Fix 7 bugs: NAS validation, ConfigDict pickling, dataset utils, SmoothedValue, error handling Apr 7, 2026
Adithya-Thonse pushed a commit that referenced this pull request Jun 12, 2026
@Adithya-Thonse
Pull request #6: Documentation for automatic mixed precision quantiza…
7899e57 
…tion

Merge in TINYML-ALGO/tinyml-tensorlab from 2026/abhijeet to main

* commit '2651f23e8d3739bd36482b65cd791e0446d9b31b':
  adding proper thresholds and using them from config.yaml file
  adding Automatic mixed precision quantization documentation under Advanced Features
  adding Automatic mixed precision quantization documentation under Advanced Features
Adithya-Thonse pushed a commit that referenced this pull request Jun 12, 2026
Pull request #6: fixes to readme and marketplace json
814316e 
Merge in TINYML-ALGO/tinyml-agent-skills from 2026/pranav_a to main

* commit 'e4bc0b462074a370f8238d6ee0353f1df6ef0cec':
  fixes to readme and marketplace json
Adithya-Thonse added a commit that referenced this pull request Jun 12, 2026
@Adithya-Thonse
Squashed 'tinyml-agent-skills/' changes from e48ef1a..de8af16d
3ccc473 
de8af16d Pull request #45: https://jira.itg.ti.com/browse/TINYML_ALGO-698
REVERT: e48ef1a Pull request #14: TINYML_ALGO-711: fixing readme
REVERT: 16fc6a6 TINYML_ALGO-711: fixing readme
REVERT: e3639d2 Pull request #13: removing pycache
REVERT: f8bb3b7 removing pycache
REVERT: dd38428 Pull request #12: restructuring agent skill
REVERT: ff02a0e restructuring agent skill
REVERT: d26c6a5 Pull request #11: fixing tiny ml name
REVERT: 640ffd3 fixing tiny ml name
REVERT: 4ee3a19 Pull request #10: 2026/pranav a
REVERT: be83fc6 minor fixes
REVERT: e3a5700 removed assets, included autoMP quant
REVERT: 1af575a Pull request #9: correcting npu devices list
REVERT: 31e9eb1 correcting npu devices list
REVERT: 59b209b Pull request #8: improving readme
REVERT: 8c3260b improving readme
REVERT: 668916f Pull request #7: improving readme
REVERT: 68686b3 improving readme
REVERT: 814316e Pull request #6: fixes to readme and marketplace json
REVERT: e4bc0b4 fixes to readme and marketplace json
REVERT: 6a64208 Pull request #5: fixes to readme
REVERT: 0f9c868 fixes to readme
REVERT: 52f95ff Pull request #4: 2026/pranav a
REVERT: 443295d fixes to readme
REVERT: 1881112 fixes to readme and marketplace json
REVERT: 229ab57 Pull request #3: 2026/pranav a
REVERT: 6519104 minor readme fix
REVERT: 38e9f9f minor readme fix
REVERT: db81f81 Pull request #2: minor readme fix
REVERT: 1c0737a minor readme fix
REVERT: 0a0c02d Pull request #1: minor readme fix
REVERT: b682335 minor readme fix
REVERT: 062eb39 Initial Commit

git-subtree-dir: tinyml-agent-skills
git-subtree-split: de8af16d9e23de3e9bda3d811a0ebdece1178260
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@musicalplatypus @t5fkg8d44d-beep