chore: release v0.3.4 - #17
Merged
Merged
Conversation
* docs(training-plan): refresh for w384 sweep, prune implemented items - add 2026-07-13 w384 backbone sweep results (5 backbones, val sfen 0.976-0.999) - collapse A/B/C/D/E design discussions to a single 実装済み summary since regression head, class weight, hand_weight=1.0, cosine+warmup, and resume have all landed - drop v3 修正結果/内訳/config tables that referenced data/ocr_v3/ files never generated - keep v3 as a compact fallback plan for the case realistic captures reveal hand-tail gaps - rewrite the next-iteration priorities around realistic evaluation and backbone selection Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs: add w384 sweep analysis with convergence metrics New document w384-sweep-analysis.md records convergence speed, epoch efficiency, and train/val gap analysis from the 384-resolution backbone sweep. Updates TRAINING_PLAN.md and docs/README.md with references. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(datasets): add HFDetectorDataset and resolution docs Add HFDetectorDataset to load detector datasets from Hugging Face Hub with per-sample bboxes. Add --hf-repo-id/--hf-config to train_detector.py. Document w224 vs w384 resolution tradeoffs with measured baselines and operational guidance. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(train): add detector training script with HF dataset support Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(datasets): enhance detector augmentation pipeline - Add asymmetric letterboxing (CropAndPad) for board context preservation - Introduce anisotropic Affine scaling (x/y independent) with adjusted transform ranges - Replace single blur with OneOf mix (GaussNoise, MotionBlur, GaussianBlur) - Increase min_visibility to 0.999 for stricter bbox inclusion in crops - Strengthen image degradation (compression, downscale) and color jitter parameters Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(eval): add realistic eval script for boardocr checkpoints Evaluate BoardOCR against ultemica/piyoshogi-eval (1k SFEN × 4 real devices). Isolates domain gap between training-set tight crops and actual device screenshots. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(detector): track best checkpoint by configurable evaluation metric Allow --best-metric argument to specify which metric selects best.pt. Checkpoints now store best_metric and best_metric_name for resume. Best model is saved to best.pt when a new best is found. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(eval): add error handling and skip counter to eval_realistic Handle missing/corrupt images gracefully with skip tracking, extract batch flush logic into a helper function, and support .env configuration loading. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(eval): add W&B integration to eval_realistic Adds W&B logging for per-(backbone, device) metrics and summary statistics. Introduces --wandb, --wandb-project, and --wandb-run-name arguments to control logging behavior. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(eval): inject wandb cf access headers via environment variable wandb-core only reads WANDB__EXTRA_HTTP_HEADERS from the environment, not from wandb.Settings, so inject CF Access credentials early before wandb imports. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * refactor(eval): require per-backbone wandb runs for 1:1 training pairing - Enforce one backbone per invocation to map eval runs 1:1 to training runs - Use backbone name as default run identifier for seamless label pairing - Add --wandb-group for clustering parallel per-backbone eval runs - Remove backbone prefix from metric keys (simplified per-run structure) - Fix summary handling for single-backbone scalar values Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs(detector): add status document for board detector implementation Reference material documenting differences between the specification (docs/ocr-model-interface.md §2) and the current BoardDetector implementation in mito_train, including input/output interface, model architecture, training results, and extension costs. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(eval): add end-to-end realistic eval with detector-predicted crops Removes crop-convention mismatch between piyoshogi-eval's bboxes and OCR training data by using detector-predicted crops. Logs per-device metrics to W&B (one run per backbone). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs: document w384 realistic evaluation and finalize backbone selection Completed piyoshogi-eval across 5 backbones × 4 devices: convnext_nano/tiny reach 99.92% (exceed val baseline), efficientnet_b1 at 99.67% (Pareto-efficient), mobilenet_v3_large at 97.92% (edge lower bound). iPad board=0% was GT bbox annotation mismatch; resolved with detector bbox preprocessing. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(export): add board_ocr export and checkpoint-aware factories MODEL_SPECS now uses factory functions that extract backbone, image_size, and hand_mode from the loaded checkpoint, preserving these settings in the exported ONNX model without requiring additional CLI arguments. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * docs: add browser WebGPU inference time outlook Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * feat(export): add ONNX export and GitHub Release pipeline Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * chore(release): bump version to 0.3.4 Detector training pipeline, realistic eval (OCR-only + end-to-end), ONNX export/release pipeline, browser inference outlook doc. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(lint): remove unused imports and empty f-string F401 in hf_detector_dataset and verify_onnx, F541 in write_release_notes. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <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.
Release v0.3.4
developをmasterに promote してプロダクション反映。What's shipping
Detector 学習パイプライン
mito_train/training/train_detector.py/HFDetectorDataset/ 強 augmentationruns/board-detector-v1が 100ep で val iou_mean 0.99+ 到達Realistic 評価
scripts/eval/eval_realistic.py(OCR 単体、GT bbox で crop)scripts/eval/eval_realistic_e2e.py(end-to-end: detector 予測 bbox で crop)mito-train-board-ocr-w384-v0.3.1-eval-realistic{,-e2e}に同期ONNX export + release パイプライン
mito_train/export/{to_onnx,to_fp16,quantize}.pyscripts/export/{export_all,verify_onnx,build_release,write_manifest,write_release_notes}.github/workflows/deployment.yml(workflow_dispatch or tagmodels-v*で発火)Doc 拡充
docs/w384-sweep-analysis.md(収束・train↔val gap + realistic 評価)docs/resolution-tradeoff.md(w224 vs w384)docs/browser-inference-outlook.md(iGPU/iOS/Android 端末別推論見積)docs/model-export.md(export パイプライン)DETECTOR_STATUS.md(仕様書との差分)Version
0.3.3 → 0.3.4(pre-1.0 rule: feat = patch)🤖 Generated with Claude Code