Spaces:
Running
Running
Commit History
9 days ago
-
feat: crown Commend API routes remix tools data processing ve istemci IP utilities eklendi Running
d411bd0 -
feat: data processing remix ve analysis backend routes ile rate limiting ve istemci IP servisleri eklendi
b65bdc6 -
feat: analysis jobs queue inference xai ve timeline modules ile unit tests eklendi
413c3b6 -
feat: analysis jobs timeline scanning inference xai ve corresponding tests eklendi
f4a412c -
feat: audio feature extraction ve vocal analyzer servisleri için music detection eklendi
8211829
17 days ago
-
feat: fastAPI backend application entrypoint ile routers ve ara yazılım eklendi
e993aad -
feat: background analysis job queue servisi ve corresponding tests eklendi
4164334 -
feat: background analysis jobs servisi ile queueing ve progress tracking eklendi
d5b9f5c -
feat: health check route için backend servisleri ve modelleri eklendi
7429d46 -
feat: health check ve audio analysis API routes eklendi
3518200 -
feat: backend entrypoint ve core analysis routes eklendi
56fa34f -
feat: background analysis jobs servisi ile queueing progress tracking ve caching eklendi
45fa2f8 -
feat: background analysis jobs servisi ile fair queueing ve caching eklendi
db51cdb -
feat: fST external API istemci için AI music detection eklendi
61387f5
19 days ago
-
feat: general analysis route için audio uploads ve YouTube URLs düzeltildi
843c1f9 -
feat: aI analysis servisleri routes detectors ve test suites eklendi
06b738c -
feat: analysis schemas XAI inference YouTube downloader ve analysis job servisleri eklendi
379d1f6
-
feat: meta-classifier servisi için stacking ensemble model fusion eklendi
f73834f -
feat: unified analysis uç noktası ve meta-classifier servisi için file ve URL processing eklendi
c52f741 -
feat: unified analysis uç noktası ile desteği için YouTube ve file uploads eklendi
063edcf -
refactor: extract Pydantic modelleri üzerinden analyze.py into analyze_schemas.py ve the core api/analyze route ile rate limiting eklendi
e6d2d3e -
feat: remix ve Multitrack API uç noktaları ile rate limiting ve Pydantic schema doğrulama eklendi
26c0314 -
feat: unified analyze uç noktası ve isolated test istemci yapılandırma eklendi
99d7ba0 -
feat: remix engine routes dream analysis servisleri ve dataset organization utilities eklendi
e49c006
-
feat: audio processing ve organization API uç noktaları ile unit testing eklendi
2eb5dd7 -
feat: youTube API servisi ve backend routing için video interaction ve comment management eklendi
8de7693
-
feat: implement automated training diagnostics, dataset bias analysis, and revision figure generation pipelines
906c392
-
feat: add reservations API with table management and email notifications
c1fe190
-
fix: extend xlim in plot_ml_vs_dl to prevent clipping of value labels
4744538
-
feat: integrate wav2vec2 score into score fusion logic
87518a8 -
fix: add wav2vec2 result handling and logging in analyze file processing
8be5b97 -
feat: add Wav2Vec2 detector service to analysis route
5529f33
-
fix: adjust confidence calculation and streamline model download logic
a55f921 -
fix: real ensemble inference, Youden threshold, DL unpickler
- inference_xai: load Youden-optimal threshold (0.4316) from training_results.json instead of hardcoded 0.5 - inference_xai: load all 11 models at startup for real-time voting (previously faked from training accuracy approximation) - inference_xai: _DLUnpickler remaps __main__.TorchSklearnWrapper so DL pkl files deserialise correctly outside training context - training_results.json: LightGBM optimal_threshold = 0.431577 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>f999d90 -
fix: update calibration and precision-recall plotting functions to load data from files
129f521 -
fix: update confusion matrix and score distribution functions to load pre-computed CV predictions
fa18330 -
feat: add script to generate publication-ready figures for AURIS paper
e8077a2 -
fix: add ffmpeg decoding fallback for audio loading in analyze_vocals
4718070
-
fix: add ffmpeg decoding fallback for unsupported audio formats in feature extraction
14fc9a2
-
fix: update evaluation to use optimal threshold and adjust BCEWithLogitsLoss for class imbalance
9b6c85d -
fix: adjust BCEWithLogitsLoss to handle class imbalance with pos_weight
5ac76c0 -
fix: add missing roc_curve metric import for evaluation
bfbdc0b -
feat: define families dictionary for candidate classifiers in _build_candidate_families function
1d86705 -
feat: enhance candidate selection with class ratio and calibrated SVC for improved model performance
57f19bf -
feat: implement optimal threshold calculation for binary predictions in train function
058eadc -
feat: add CalibratedClassifierCV and roc_curve imports for enhanced model evaluation
27cd744
-
feat: add TorchSklearnWrapper for PyTorch model compatibility with sklearn and enhance final model training process
bb33480
-
Add Gradio dependency to requirements.txt with version constraints
20fe6c3 -
feat: train all models on full data and save their paths
94ed8e9 -
Refactor code structure for improved readability and maintainability
c0f3b71