Atomizer

Antoine/Atomizer

Fork 0

Commit Graph

Author	SHA1	Message	Date
Anto01	b4c0831230	fix: Remove redundant save() call that overwrote NX expression updates Critical bug fix for LLM mode optimization: Problem: - NXParameterUpdater.update_expressions() uses NX journal to import expressions (default use_nx_import=True) - The NX journal directly updates the PRT file on disk and saves it - But then run_optimization.py was calling updater.save() afterwards - save() writes self.content (loaded at initialization) back to file - This overwrote the NX journal changes with stale binary content! Result: All optimization trials produced identical FEM results because the model was never actually updated. Fixes: 1. Removed updater.save() call from model_updater closure in run_optimization.py 2. Added theSession.Parts.CloseAll() in import_expressions.py to ensure changes are flushed and file is released 3. Fixed test_phase_3_2_e2e.py variable name (best_trial_file → results_file) Testing: Verified expressions persist to disk correctly with standalone test. Next step: Address remaining issue where FEM results are still identical (likely solve journal not reloading updated PRT).	2025-11-17 21:24:02 -05:00
Anto01	90a9e020d8	feat: Complete Phase 3.1 - Extractor Orchestration & End-to-End Automation Phase 3.1 completes the ZERO-MANUAL-CODING automation pipeline by integrating all phases into a seamless workflow from natural language request to final objective value. Key Features: - ExtractorOrchestrator integrates Phase 2.7 LLM + Phase 3.0 Research Agent - Automatic extractor generation from LLM workflow output - Dynamic loading and execution on real OP2 files - Smart parameter filtering per extraction pattern type - Multi-extractor support in single workflow - Complete end-to-end test passed on real bracket OP2 Complete Automation Pipeline: User Natural Language Request ↓ Phase 2.7 LLM Analysis ↓ Phase 3.1 Orchestrator ↓ Phase 3.0 Research Agent (auto OP2 code gen) ↓ Generated Extractor Modules ↓ Dynamic Execution on Real OP2 ↓ Phase 2.8 Inline Calculations ↓ Phase 2.9 Post-Processing Hooks ↓ Final Objective → Optuna Test Results: - Generated displacement extractor: PASSED - Executed on bracket OP2: PASSED - Extracted max_displacement: 0.361783mm at node 91 - Calculated normalized objective: 0.072357 - Multi-extractor generation: PASSED New Files: - optimization_engine/extractor_orchestrator.py (380+ lines) - tests/test_phase_3_1_integration.py (200+ lines) - docs/SESSION_SUMMARY_PHASE_3_1.md (comprehensive documentation) - optimization_engine/result_extractors/generated/ (auto-generated extractors) Modified Files: - README.md - Added Phase 3.1 completion status ZERO MANUAL CODING - Complete automation achieved! Generated with Claude Code Co-Authored-By: Claude <noreply@anthropic.com>	2025-11-16 19:39:04 -05:00

Author

SHA1

Message

Date

Anto01

b4c0831230

fix: Remove redundant save() call that overwrote NX expression updates

Critical bug fix for LLM mode optimization:

**Problem**:
- NXParameterUpdater.update_expressions() uses NX journal to import expressions (default use_nx_import=True)
- The NX journal directly updates the PRT file on disk and saves it
- But then run_optimization.py was calling updater.save() afterwards
- save() writes self.content (loaded at initialization) back to file
- This overwrote the NX journal changes with stale binary content!

**Result**: All optimization trials produced identical FEM results because the model was never actually updated.

**Fixes**:
1. Removed updater.save() call from model_updater closure in run_optimization.py
2. Added theSession.Parts.CloseAll() in import_expressions.py to ensure changes are flushed and file is released
3. Fixed test_phase_3_2_e2e.py variable name (best_trial_file → results_file)

**Testing**: Verified expressions persist to disk correctly with standalone test.

Next step: Address remaining issue where FEM results are still identical (likely solve journal not reloading updated PRT).

2025-11-17 21:24:02 -05:00

Anto01

90a9e020d8

feat: Complete Phase 3.1 - Extractor Orchestration & End-to-End Automation

Phase 3.1 completes the ZERO-MANUAL-CODING automation pipeline by
integrating all phases into a seamless workflow from natural language
request to final objective value.

Key Features:
- ExtractorOrchestrator integrates Phase 2.7 LLM + Phase 3.0 Research Agent
- Automatic extractor generation from LLM workflow output
- Dynamic loading and execution on real OP2 files
- Smart parameter filtering per extraction pattern type
- Multi-extractor support in single workflow
- Complete end-to-end test passed on real bracket OP2

Complete Automation Pipeline:
  User Natural Language Request
      ↓
  Phase 2.7 LLM Analysis
      ↓
  Phase 3.1 Orchestrator
      ↓
  Phase 3.0 Research Agent (auto OP2 code gen)
      ↓
  Generated Extractor Modules
      ↓
  Dynamic Execution on Real OP2
      ↓
  Phase 2.8 Inline Calculations
      ↓
  Phase 2.9 Post-Processing Hooks
      ↓
  Final Objective → Optuna

Test Results:
- Generated displacement extractor: PASSED
- Executed on bracket OP2: PASSED
- Extracted max_displacement: 0.361783mm at node 91
- Calculated normalized objective: 0.072357
- Multi-extractor generation: PASSED

New Files:
- optimization_engine/extractor_orchestrator.py (380+ lines)
- tests/test_phase_3_1_integration.py (200+ lines)
- docs/SESSION_SUMMARY_PHASE_3_1.md (comprehensive documentation)
- optimization_engine/result_extractors/generated/ (auto-generated extractors)

Modified Files:
- README.md - Added Phase 3.1 completion status

ZERO MANUAL CODING - Complete automation achieved!

Generated with Claude Code

Co-Authored-By: Claude <noreply@anthropic.com>

2025-11-16 19:39:04 -05:00

2 Commits