Changelog

All notable changes to Entrenar will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

0.5.6 - 2026-01-22

Changed

Code Organization (Technical Debt Reduction)

  • Exploded 16 large modules (>1000 lines) into directory module structures
    • storage/sqlite/backend.rs → backend/ (sqlite_backend, state, tests)
    • config/train.rs → train/ (loader, batches, arrow, demo, tests)
    • eval/evaluator.rs → evaluator/ (metric, config, result, leaderboard, kfold, model_evaluator, tests)
    • hf_pipeline/fetcher.rs → fetcher/ (types, options, hf_fetcher, tests)
    • efficiency/device.rs → device/ (simd, cpu, gpu, tpu, apple, compute, tests)
    • monitor/llm.rs → llm/ (error, metrics, prompt, eval_result, traits, stats, memory_evaluator, heuristics, tests)
    • monitor/inference/provenance.rs → provenance/ (node, edge, graph, attack, reconstructor, tests)
    • citl/pattern_store.rs → pattern_store/ (chunk_id, fix_pattern, suggestion, config, store, data, tests)
    • storage/registry.rs → registry/ (stage, version, comparison, transition, policy, error, traits, memory)
    • citl/trainer.rs → trainer/ (span, trace, outcome, correlation, config, stats, citl)
    • tokenizer/mod.rs → separate files (error, config, traits, bpe, char, hf)
    • monitor/wasm.rs → wasm/ (collector, options, dashboard, utils)
    • Plus 4 from previous release: preflight, cloud, code_gan, manifest

Quality

  • 3710 tests passing (100% success rate)
  • 96.66% code coverage (exceeds 95% target)
  • PMAT compliance: COMPLIANT
  • File health: Only 3 files >1000 lines (main.rs + 2 test files)
  • All API compatibility maintained via re-exports

Dependencies

  • Updated PAIML stack dependencies

0.1.0 - 2025-11-21

Added

Core Framework

  • Autograd Engine - Tape-based automatic differentiation with backward propagation
    • Tensor abstraction with gradient tracking
    • BackwardOp trait for custom operations
    • Attention, matmul, softmax, layer norm operations
    • Property-based gradient checking (200K+ iterations)

Optimizers

  • SGD with momentum support
  • Adam optimizer with bias correction
  • AdamW with decoupled weight decay
  • Gradient clipping via L2 norm
  • Learning rate scheduling (Cosine, Linear)
  • SIMD acceleration for parameter updates via Trueno
  • Convergence property tests for all optimizers

LoRA & QLoRA

  • LoRA layers with configurable rank and alpha
  • QLoRA with 4-bit quantized base weights
  • Adapter management (save/load separately from base model)
  • Memory benchmarks showing 4× reduction with QLoRA
  • Gradient flow tests ensuring proper backpropagation

Quantization

  • QAT (Quantization-Aware Training) with fake quantize
  • PTQ (Post-Training Quantization) with calibration
  • 4-bit and 8-bit quantization support
  • Symmetric and asymmetric quantization modes
  • Per-channel and per-tensor quantization
  • Compression ratio validation and accuracy degradation tests

Model Merging (Arcee Methods)

  • TIES (Task Inference via Elimination and Sign voting)
  • DARE (Drop And REscale with Bernoulli masking)
  • SLERP (Spherical Linear intERPolation)
  • Property tests for permutation invariance
  • Multi-model ensemble support

Knowledge Distillation

  • Temperature-scaled KL divergence loss
  • Multi-teacher ensemble distillation
  • Progressive layer-wise distillation
  • 44 distillation tests including 13 property tests
  • Temperature smoothing validation

Declarative Configuration

  • YAML-based training configuration (Ludwig-style)
  • Schema validation with comprehensive error messages
  • Auto-inference of feature types from data
  • Single-command training via train_from_yaml()
  • Builder pattern for optimizers and models from config

Training Loop

  • High-level Trainer abstraction
  • Batch processing with configurable batch size
  • Metrics tracking (loss history, learning rates, steps)
  • Gradient clipping integration
  • Learning rate scheduling during training
  • train_step() and train_epoch() methods

Model I/O

  • Save/load models with multiple formats
    • JSON (pretty-printed or compact)
    • YAML for human-readable configs
    • Placeholder for GGUF (future Realizar integration)
  • ModelMetadata with custom fields
  • Round-trip integrity validation
  • Automatic format detection from file extension

Testing & Quality

  • 258 tests passing (100% success rate)
    • Unit tests for all modules
    • Integration tests for end-to-end workflows
    • Property-based tests (200K+ iterations)
    • Gradient correctness validation
    • Round-trip serialization tests
  • 0 clippy warnings (strict mode)
  • 0 TODOs remaining in codebase
  • 55 Rust source files with full documentation

Examples

  • training_loop.rs - Demonstrates Trainer API
  • model_io.rs - Save/load workflow
  • train_from_yaml_example.rs - Declarative training
  • distillation.rs - Knowledge distillation
  • merge_models.rs - Model merging methods
  • train_from_yaml.rs - YAML configuration
  • Plus LLAMA2 examples (train, finetune-lora, finetune-qlora, memory-benchmarks)

Documentation

  • Comprehensive API documentation for all public modules
  • README with quick start guide
  • Specification documents for all major components
  • Example configurations (config.yaml)

Dependencies

  • trueno 0.4.1 - SIMD-accelerated compute engine
  • ndarray 0.16 - N-dimensional arrays
  • serde 1.0 - Serialization framework
  • thiserror 2.0 - Error handling
  • proptest 1.4 - Property-based testing (dev)
  • tempfile 3.8 - Testing utilities (dev)

Notes

  • This is the initial release of Entrenar
  • GGUF loading requires future Realizar integration
  • Real data loading (Parquet/CSV) to be added
  • Performance benchmarks to be published