mirror of
				https://github.com/ggml-org/llama.cpp.git
				synced 2025-11-03 09:22:01 +00:00 
			
		
		
		
	imatrix : use GGUF to store imatrix data
This commit is contained in:
		@@ -167,6 +167,12 @@ class Keys:
 | 
			
		||||
        TYPE       = "adapter.type"
 | 
			
		||||
        LORA_ALPHA = "adapter.lora.alpha"
 | 
			
		||||
 | 
			
		||||
    class IMatrix:
 | 
			
		||||
        CHUNK_COUNT = "imatrix.chunk_count"
 | 
			
		||||
        CHUNK_SIZE  = "imatrix.chunk_size"
 | 
			
		||||
        DATASET     = "imatrix.dataset"
 | 
			
		||||
 | 
			
		||||
 | 
			
		||||
#
 | 
			
		||||
# recommended mapping of model tensor names for storage in gguf
 | 
			
		||||
#
 | 
			
		||||
@@ -175,6 +181,7 @@ class Keys:
 | 
			
		||||
class GGUFType:
 | 
			
		||||
    MODEL   = "model"
 | 
			
		||||
    ADAPTER = "adapter"
 | 
			
		||||
    IMATRIX = "imatrix"
 | 
			
		||||
 | 
			
		||||
 | 
			
		||||
class MODEL_ARCH(IntEnum):
 | 
			
		||||
 
 | 
			
		||||
		Reference in New Issue
	
	Block a user