MCPcopy Create free account
hub / github.com/andrewkchan/deepseek.cpp / convert.py

File convert.py

convert.py:None–None  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

1# Converts a model consisting of a huggingface config.json, tokenizer.json, and .safetensors weights into a .yalm file,
2# which:
3# - Normalizes the config to a common format in the header
4# - Combines any safetensors shards

Callers

nothing calls this directly

Calls 8

BlockQuantClass · 0.85
KQuantClass · 0.85
MetadataClass · 0.85
load_tokensFunction · 0.85
load_weightsFunction · 0.85
loadMethod · 0.80
to_dictMethod · 0.80
parse_argsMethod · 0.45

Tested by

no test coverage detected