mirror of https://github.com/google/gemma.cpp.git
8d0882b966
Weight handling: - new ModelStore2 supports both pre-2025 multi-file and single-file formats - simpler ForEachTensor with TensorArgs - tensors are constructed with their full suffixed name I/O: - support mmap and stride - Simplified SbsWriter, single insert(); add SbsReader Misc: - kMockTokenizer: allow creating with unavailable tokenizer - configs.h: Simpler enum validity checks via kSentinel - matmul.h: remove unused enable_bind (now in allocator.h) - tensor_info: single TensorInfoRegistry class, rename from tensor_index.h Frontends: - Replace Allocate/CreateGemma with ctor(LoaderArgs, MatMulEnv&) - Deduce model/weight type, remove --model and parsing - Replace most common.h includes with configs.h - Remove --compressed_weights, use --weights instead - Remove ModelInfo, replaced by ModelConfig. Backprop: - Reduce max loss, remove backward_scalar_test (timeout) - Update thresholds because new RandInit changes rng eval order and thus numerics PiperOrigin-RevId: 755317484 |
||
|---|---|---|
| .. | ||
| bindings | ||
| evals | ||
| instantiations | ||
| activations.h | ||
| common.cc | ||
| common.h | ||
| configs.cc | ||
| configs.h | ||
| configs_test.cc | ||
| gemma-inl.h | ||
| gemma.cc | ||
| gemma.h | ||
| gemma_args.h | ||
| kv_cache.cc | ||
| kv_cache.h | ||
| model_store.cc | ||
| model_store.h | ||
| run.cc | ||
| tensor_info.cc | ||
| tensor_info.h | ||
| tensor_info_test.cc | ||
| tokenizer.cc | ||
| tokenizer.h | ||
| weights.cc | ||
| weights.h | ||