Add quantize-stats command for testing quantization (#728)

Command that calculates some statistics over the errors introduced by
quantization, like mean square error, max error and some percentile errors for layer
weights. Should be useful for testing quantization improvements.

Exposes some internal state from ggml and llama for testing

1 files changed, 5 insertions, 0 deletions

diff --git a/llama.cpp b/llama.cpp
index 978327a5..fc6f43af 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -1852,3 +1852,8 @@ const char * llama_print_system_info(void) {
 
     return s.c_str();
 }
+
+// For internal test use
+std::unordered_map<std::string, struct ggml_tensor *>& llama_internal_get_tensor_map(struct llama_context * ctx) {
+    return ctx->model.tensors;
+}