trainingsample 0.1.0__tar.gz → 0.2.1__tar.gz

trainingsample-0.2.1/.envrc

@@ -0,0 +1,15 @@
+# Environment variables for OpenCV Rust bindings
+export DYLD_LIBRARY_PATH="/opt/homebrew/Cellar/llvm/21.1.1/lib:$DYLD_LIBRARY_PATH"
+export LIBCLANG_PATH="/opt/homebrew/Cellar/llvm/21.1.1/lib"
+export LLVM_CONFIG_PATH="/opt/homebrew/Cellar/llvm/21.1.1/bin/llvm-config"
+
+# OpenCV environment variables
+export OPENCV_LINK_LIBS="opencv_calib3d,opencv_core,opencv_dnn,opencv_features2d,opencv_flann,opencv_highgui,opencv_imgcodecs,opencv_imgproc,opencv_ml,opencv_objdetect,opencv_photo,opencv_stitching,opencv_video,opencv_videoio"
+export OPENCV_LINK_PATHS="/opt/homebrew/lib"
+export OPENCV_INCLUDE_PATHS="/opt/homebrew/include/opencv4"
+
+# Rust build optimization
+export RUSTFLAGS="-C target-cpu=native"
+
+# macOS deployment target - match current system
+export MACOSX_DEPLOYMENT_TARGET="$(sw_vers -productVersion | cut -d. -f1,2)"

trainingsample-0.2.1/.github/workflows/ci.yml

@@ -0,0 +1,234 @@
+name: CI
+
+on:
+  push:
+    branches: [ main, develop ]
+  pull_request:
+    branches: [ main ]
+
+env:
+  CARGO_TERM_COLOR: always
+
+jobs:
+  rust-tests:
+    name: Rust Tests
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        rust: [stable, beta]
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Install OpenCV and LLVM dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y pkg-config libopencv-dev clang libclang-dev llvm-dev
+        # Install LLVM 18 which has the proper libclang.so symlink
+        sudo apt-get install -y llvm-18 llvm-18-dev llvm-18-tools libclang1-18 libclang-cpp18
+        # Set environment variables for LLVM 18
+        echo "LIBCLANG_PATH=/usr/lib/llvm-18/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/usr/bin/llvm-config-18" >> $GITHUB_ENV
+        echo "LD_LIBRARY_PATH=/usr/lib/llvm-18/lib:$LD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Verify the key files exist
+        ls -la /usr/lib/llvm-18/lib/libclang.so
+        which llvm-config-18
+
+    - name: Install Rust toolchain
+      uses: dtolnay/rust-toolchain@master
+      with:
+        toolchain: ${{ matrix.rust }}
+        components: rustfmt, clippy
+
+    - name: Cache cargo registry
+      uses: actions/cache@v4
+      with:
+        path: ~/.cargo/registry
+        key: ${{ runner.os }}-cargo-registry-${{ hashFiles('**/Cargo.lock') }}
+
+    - name: Cache cargo index
+      uses: actions/cache@v4
+      with:
+        path: ~/.cargo/git
+        key: ${{ runner.os }}-cargo-index-${{ hashFiles('**/Cargo.lock') }}
+
+    - name: Cache cargo target
+      uses: actions/cache@v4
+      with:
+        path: target
+        key: ${{ runner.os }}-cargo-build-target-${{ hashFiles('**/Cargo.lock') }}
+
+    - name: Check formatting
+      run: cargo fmt --all -- --check
+
+    - name: Run clippy
+      run: cargo clippy --all-targets --no-default-features --features python-bindings,simd,opencv -- -D warnings
+
+    - name: Run tests (no Python bindings)
+      run: cargo test --no-default-features --features simd,opencv
+
+    - name: Compile tests (Linux compatible features)
+      run: cargo test --no-default-features --features python-bindings,simd,opencv --no-run
+
+  python-tests:
+    name: Python Tests
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.11", "3.12"]
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Install OpenCV and LLVM dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y pkg-config libopencv-dev clang libclang-dev llvm-dev
+        # Install LLVM 18 which has the proper libclang.so symlink
+        sudo apt-get install -y llvm-18 llvm-18-dev llvm-18-tools libclang1-18 libclang-cpp18
+        # Set environment variables for LLVM 18
+        echo "LIBCLANG_PATH=/usr/lib/llvm-18/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/usr/bin/llvm-config-18" >> $GITHUB_ENV
+        echo "LD_LIBRARY_PATH=/usr/lib/llvm-18/lib:$LD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Verify the key files exist
+        ls -la /usr/lib/llvm-18/lib/libclang.so
+        which llvm-config-18
+
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v4
+      with:
+        python-version: ${{ matrix.python-version }}
+
+    - name: Install Rust toolchain
+      uses: dtolnay/rust-toolchain@stable
+
+    - name: Install maturin (Linux/macOS)
+      run: pip install maturin[patchelf]
+
+    - name: Build Python package (Linux)
+      if: runner.os == 'Linux'
+      run: |
+        mkdir -p dist
+        maturin build --release --out dist --no-default-features --features python-bindings,simd,opencv
+
+    - name: Build Python package (macOS)
+      if: runner.os == 'macOS'
+      run: |
+        mkdir -p dist
+        maturin build --release --out dist --no-default-features --features python-bindings,simd,metal
+
+    - name: Install built package (Linux/macOS)
+      run: pip install dist/*.whl
+
+    - name: Install test dependencies
+      run: pip install pytest pytest-benchmark numpy pillow opencv-python
+
+    - name: Run Python tests
+      run: pytest tests/ -v
+
+  pre-commit:
+    name: Pre-commit hooks
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Install OpenCV and LLVM dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y pkg-config libopencv-dev clang libclang-dev llvm-dev
+        # Install LLVM 18 which has the proper libclang.so symlink
+        sudo apt-get install -y llvm-18 llvm-18-dev llvm-18-tools libclang1-18 libclang-cpp18
+        # Set environment variables for LLVM 18
+        echo "LIBCLANG_PATH=/usr/lib/llvm-18/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/usr/bin/llvm-config-18" >> $GITHUB_ENV
+        echo "LD_LIBRARY_PATH=/usr/lib/llvm-18/lib:$LD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Verify the key files exist
+        ls -la /usr/lib/llvm-18/lib/libclang.so
+        which llvm-config-18
+
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: 3.11
+
+    - name: Install Rust toolchain
+      uses: dtolnay/rust-toolchain@stable
+      with:
+        components: rustfmt, clippy
+
+    - name: Install cargo-audit
+      run: cargo install cargo-audit
+
+    - name: Install pre-commit
+      run: pip install pre-commit
+
+    - name: Run pre-commit
+      run: pre-commit run --all-files
+
+  build-wheels:
+    name: Build wheels
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-latest, macos-latest]
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Install OpenCV and LLVM dependencies (Ubuntu)
+      if: runner.os == 'Linux'
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y pkg-config libopencv-dev clang libclang-dev llvm-dev
+        # Install LLVM 18 which has the proper libclang.so symlink
+        sudo apt-get install -y llvm-18 llvm-18-dev llvm-18-tools libclang1-18 libclang-cpp18
+        # Set environment variables for LLVM 18
+        echo "LIBCLANG_PATH=/usr/lib/llvm-18/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/usr/bin/llvm-config-18" >> $GITHUB_ENV
+        echo "LD_LIBRARY_PATH=/usr/lib/llvm-18/lib:$LD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Verify the key files exist
+        ls -la /usr/lib/llvm-18/lib/libclang.so
+        which llvm-config-18
+
+
+
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: 3.11
+
+    - name: Install Rust toolchain
+      uses: dtolnay/rust-toolchain@stable
+
+    - name: Install maturin (Linux/macOS)
+      run: pip install maturin[patchelf]
+
+    - name: Install OpenCV (macOS)
+      if: runner.os == 'macOS'
+      run: |
+        # Install OpenCV and LLVM via Homebrew
+        brew install opencv llvm
+        # Set up LLVM environment for opencv-rust bindings (keg-only installation)
+        echo "LIBCLANG_PATH=/opt/homebrew/opt/llvm/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/opt/homebrew/opt/llvm/bin/llvm-config" >> $GITHUB_ENV
+        echo "DYLD_LIBRARY_PATH=/opt/homebrew/opt/llvm/lib:$DYLD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Set up OpenCV environment
+        echo "OPENCV_LINK_LIBS=opencv_calib3d,opencv_core,opencv_dnn,opencv_features2d,opencv_flann,opencv_highgui,opencv_imgcodecs,opencv_imgproc,opencv_ml,opencv_objdetect,opencv_photo,opencv_stitching,opencv_video,opencv_videoio" >> $GITHUB_ENV
+        echo "OPENCV_LINK_PATHS=/opt/homebrew/lib" >> $GITHUB_ENV
+        echo "OPENCV_INCLUDE_PATHS=/opt/homebrew/include/opencv4" >> $GITHUB_ENV
+
+    - name: Build wheels (Linux)
+      if: runner.os == 'Linux'
+      run: maturin build --release --out dist --no-default-features --features python-bindings,simd,opencv
+
+    - name: Build wheels (macOS)
+      if: runner.os == 'macOS'
+      run: maturin build --release --out dist --no-default-features --features python-bindings,simd,opencv,metal
+
+
+    - name: Upload wheels
+      uses: actions/upload-artifact@v4
+      with:
+        name: wheels-${{ matrix.os }}
+        path: dist/

{trainingsample-0.1.0 → trainingsample-0.2.1}/.github/workflows/publish.yml

@@ -17,16 +17,25 @@ jobs:
       with:
         python-version: '3.11'
 
-    - name: Install FFmpeg dependencies
+    - name: Install OpenCV and LLVM dependencies
       run: |
         sudo apt-get update
-        sudo apt-get install -y pkg-config libavutil-dev libavformat-dev libavcodec-dev libavdevice-dev libavfilter-dev libswscale-dev libswresample-dev
+        sudo apt-get install -y pkg-config libopencv-dev clang libclang-dev llvm-dev
+        # Install LLVM 18 which has the proper libclang.so symlink
+        sudo apt-get install -y llvm-18 llvm-18-dev llvm-18-tools libclang1-18 libclang-cpp18
+        # Set environment variables for LLVM 18
+        echo "LIBCLANG_PATH=/usr/lib/llvm-18/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/usr/bin/llvm-config-18" >> $GITHUB_ENV
+        echo "LD_LIBRARY_PATH=/usr/lib/llvm-18/lib:$LD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Verify the key files exist
+        ls -la /usr/lib/llvm-18/lib/libclang.so
+        which llvm-config-18
 
     - name: Build wheels
       uses: PyO3/maturin-action@v1
       with:
         target: ${{ matrix.target }}
-        args: --release --out dist --find-interpreter --zig
+        args: --release --out dist --find-interpreter --zig --no-default-features --features python-bindings,simd,opencv
         sccache: 'true'
         manylinux: auto
       env:
@@ -39,31 +48,6 @@ jobs:
         name: wheels-linux-${{ matrix.target }}
         path: dist
 
-  windows:
-    runs-on: windows-latest
-    strategy:
-      matrix:
-        target: [x64]
-    steps:
-    - uses: actions/checkout@v4
-    - uses: actions/setup-python@v5
-      with:
-        python-version: '3.11'
-        architecture: ${{ matrix.target }}
-    - name: Build wheels
-      uses: PyO3/maturin-action@v1
-      with:
-        target: ${{ matrix.target }}
-        args: --release --out dist --find-interpreter --no-default-features --features python-bindings
-        sccache: 'true'
-      env:
-        # Use stable ABI for forward compatibility
-        PYO3_USE_ABI3_FORWARD_COMPATIBILITY: '1'
-    - name: Upload wheels
-      uses: actions/upload-artifact@v4
-      with:
-        name: wheels-windows-${{ matrix.target }}
-        path: dist
 
   macos:
     runs-on: macos-latest
@@ -76,11 +60,24 @@ jobs:
       with:
         python-version: '3.11'
 
+    - name: Install OpenCV (macOS)
+      run: |
+        # Install OpenCV and LLVM via Homebrew
+        brew install opencv llvm
+        # Set up LLVM environment for opencv-rust bindings (keg-only installation)
+        echo "LIBCLANG_PATH=/opt/homebrew/opt/llvm/lib" >> $GITHUB_ENV
+        echo "LLVM_CONFIG_PATH=/opt/homebrew/opt/llvm/bin/llvm-config" >> $GITHUB_ENV
+        echo "DYLD_LIBRARY_PATH=/opt/homebrew/opt/llvm/lib:$DYLD_LIBRARY_PATH" >> $GITHUB_ENV
+        # Set up OpenCV environment
+        echo "OPENCV_LINK_LIBS=opencv_calib3d,opencv_core,opencv_dnn,opencv_features2d,opencv_flann,opencv_highgui,opencv_imgcodecs,opencv_imgproc,opencv_ml,opencv_objdetect,opencv_photo,opencv_stitching,opencv_video,opencv_videoio" >> $GITHUB_ENV
+        echo "OPENCV_LINK_PATHS=/opt/homebrew/lib" >> $GITHUB_ENV
+        echo "OPENCV_INCLUDE_PATHS=/opt/homebrew/include/opencv4" >> $GITHUB_ENV
+
     - name: Build wheels
       uses: PyO3/maturin-action@v1
       with:
         target: ${{ matrix.target }}
-        args: --release --out dist --find-interpreter --no-default-features --features python-bindings
+        args: --release --out dist --find-interpreter --no-default-features --features python-bindings,simd,opencv,metal
         sccache: 'true'
       env:
         # Use stable ABI for forward compatibility
@@ -109,7 +106,7 @@ jobs:
   release:
     name: Release
     runs-on: ubuntu-latest
-    needs: [linux, windows, macos, sdist]
+    needs: [linux, macos, sdist]
     steps:
     - uses: actions/download-artifact@v4
       with:

{trainingsample-0.1.0 → trainingsample-0.2.1}/.pre-commit-config.yaml

@@ -13,14 +13,14 @@ repos:
       - id: cargo-check
         name: Rust check
         description: Check Rust code with cargo check
-        entry: cargo check --all-targets --all-features
+        entry: bash -c 'if [[ "$OSTYPE" == "darwin"* ]]; then cargo check --all-targets --features simd,opencv,python-bindings,metal; else cargo check --all-targets --features simd,opencv,python-bindings; fi'
         language: system
         files: \.rs$
         pass_filenames: false
       - id: clippy
         name: Rust clippy
         description: Lint Rust code with clippy
-        entry: cargo clippy --all-targets --all-features
+        entry: bash -c 'if [[ "$OSTYPE" == "darwin"* ]]; then cargo clippy --all-targets --features simd,opencv,python-bindings,metal; else cargo clippy --all-targets --features simd,opencv,python-bindings; fi'
         language: system
         files: \.rs$
         pass_filenames: false
@@ -72,7 +72,7 @@ repos:
       - id: rust-test
         name: Rust tests
         description: Run Rust tests
-        entry: cargo test --features video
+        entry: bash -c 'if [[ "$OSTYPE" == "darwin"* ]]; then cargo test --features simd,opencv,metal -- --test-threads=1; else cargo test --features simd,opencv -- --test-threads=1; fi'
         language: system
         files: \.rs$
         pass_filenames: false
@@ -89,3 +89,15 @@ repos:
         files: \.(py|rs)$
         pass_filenames: false
         stages: [pre-commit]
+
+  # Competitive benchmarks (CI only - too heavy for local precommit)
+  - repo: local
+    hooks:
+      - id: competitive-benchmarks
+        name: Competitive performance benchmarks
+        description: Run competitive benchmarks against OpenCV/NumPy
+        entry: bash -c 'if [[ "$CI" == "true" ]] && ([ -d "target/wheels" ] || python -c "import trainingsample" 2>/dev/null); then pytest tests/test_competitive_benchmarks.py -v --tb=short; else echo "Skipping competitive benchmarks (not in CI or bindings not available)"; fi'
+        language: system
+        files: \.(py|rs)$
+        pass_filenames: false
+        stages: [pre-commit]

trainingsample-0.2.1/BENCHMARKS.md

@@ -0,0 +1,112 @@
+# Competitive Performance Benchmarks
+
+This library includes comprehensive benchmarks against industry-standard libraries (OpenCV, NumPy) to ensure competitive performance for real-world SFT (Supervised Fine-Tuning) workloads.
+
+## Benchmark Categories
+
+### 🖼️ High-Resolution Image Processing
+
+**Target Workload**: 5120×5120 → 1024×1024 image processing pipeline
+- **Input**: 5120×5120×3 images (26.2M pixels, ~78MB each)
+- **Pipeline**: Center crop → Resize → Luminance calculation
+- **Batch sizes**: 2-4 images (memory constrained)
+
+### 📊 Performance Targets
+
+| Operation | Input Size | Target Performance | Baseline |
+|-----------|------------|-------------------|----------|
+| **Resize** | 5120×5120 → 1024×1024 | Match OpenCV bilinear | `cv2.resize()` |
+| **Center Crop** | 5120×5120 → 2048×2048 | Match/exceed NumPy | Array slicing |
+| **Luminance** | 1024×1024 | 1.5x+ faster than NumPy | Vectorized math |
+| **Full Pipeline** | 5120×5120 → 1024×1024 | >0.5 images/sec | Combined ops |
+
+### 🎯 Quality Targets
+
+- **Resize Quality**: PSNR >30dB vs OpenCV (excellent similarity)
+- **Crop Accuracy**: Bit-exact match with NumPy center crop
+- **Luminance Precision**: <0.1 difference vs NumPy reference
+
+## Running Benchmarks
+
+### Local Development
+```bash
+# Install dependencies
+pip install opencv-python pytest-benchmark psutil
+
+# Build with optimizations
+maturin develop --release --features "python-bindings,simd"
+
+# Run competitive benchmarks
+./scripts/run_competitive_benchmarks.sh
+```
+
+### CI/CD Integration
+
+Benchmarks run automatically in CI for:
+- **Pull requests**: Performance regression detection
+- **Main branch**: Performance tracking over time
+- **Weekly schedule**: Long-term performance monitoring
+
+## Benchmark Architecture
+
+### Memory Efficiency
+- Monitors RSS memory usage vs OpenCV
+- Tests batch processing memory scaling
+- Validates no memory leaks in pipelines
+
+### SIMD Optimization Validation
+- Compares SIMD-enabled vs scalar fallback performance
+- Tests x86-64 AVX2/AVX-512 and ARM64 NEON paths
+- Validates CPU feature detection accuracy
+
+### Real-World Scenarios
+- **SFT Data Processing**: High-res → training resolution pipeline
+- **Batch Processing**: Multiple images with different operations
+- **Memory Constraints**: Large images with limited RAM
+
+## Performance Philosophy
+
+### Why These Benchmarks Matter
+
+1. **Real-World Relevance**: SFT workloads use 5120×5120+ images, not toy 224×224
+2. **Competitive Pressure**: OpenCV and NumPy are highly optimized incumbents
+3. **User Experience**: Poor performance = adoption barriers
+4. **Resource Efficiency**: Training infrastructure costs scale with throughput
+
+### Performance vs Quality Tradeoffs
+
+- **Resize**: Bilinear interpolation for speed, good quality balance
+- **SIMD**: Aggressive optimization while maintaining numerical accuracy
+- **Memory**: Batch processing for throughput vs memory pressure balance
+
+## Interpreting Results
+
+### Good Performance Indicators
+- ✅ Resize: 1-2 images/sec for 5120×5120 → 1024×1024
+- ✅ Crop: 10+ images/sec for 5120×5120 → 2048×2048
+- ✅ Luminance: 1.5x+ faster than NumPy with SIMD
+- ✅ Pipeline: >0.5 complete transformations/sec
+
+### Red Flags
+- ❌ Slower than OpenCV resize (indicates poor SIMD utilization)
+- ❌ Slower than NumPy crop (indicates unnecessary overhead)
+- ❌ Memory usage >2x OpenCV (indicates memory leaks/inefficiency)
+- ❌ Quality degradation (PSNR <30dB vs reference)
+
+## Future Enhancements
+
+### Planned Improvements
+- GPU acceleration benchmarks (Metal/CUDA)
+- More interpolation methods (bicubic, lanczos)
+- Video processing pipeline benchmarks
+- Multi-threaded batch processing optimization
+
+### Performance Tracking
+- Historical performance database
+- Regression detection and alerting
+- Performance comparison across different hardware configurations
+- Automated performance optimization recommendations
+
+---
+
+**Goal**: Be the fastest, highest-quality image processing library for ML/SFT workloads while maintaining competitive memory usage and numerical accuracy.