vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

GGUF: C++ refactor, backend support, misc fixes (#11030)

Browse source 
* GGUF: C++ refactor, backend support, misc fixes

remove ggml_tensor.backend

update CODEOWNERS [no ci]

remove gguf_get_data from API

revise GGUF API data types
This commit is contained in:
Johannes Gäßler 2025-01-07 18:01:58 +01:00 • committed by GitHub
parent 017cc5f446
commit 53ff6b9b9f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
21 changed files with 1795 additions and 1627 deletions
Download patch file Download diff file Expand all files Collapse all files

14
examples/gguf-split/gguf-split.cpp
View file

@ -1,16 +1,18 @@
#include "ggml.h"
#include "gguf.h"
#include "llama.h"
#include "common.h"
#include <algorithm>
#include <cinttypes>
#include <climits>
#include <cstdio>
#include <cstdlib>
#include <stdexcept>
#include <cstring>
#include <fstream>
#include <string>
#include <vector>
#include <climits>
#include <cstdio>
#include <cstring>
#include <stdexcept>
#if defined(_WIN32)
#include <windows.h>
@ -296,7 +298,7 @@ struct split_strategy {
total_size += ggml_nbytes(t);
}
total_size = total_size / 1000 / 1000; // convert to megabytes
printf("split %05d: n_tensors = %d, total_size = %zuM\n", i_split + 1, gguf_get_n_tensors(ctx_out), total_size);
printf("split %05d: n_tensors = %" PRIi64 ", total_size = %zuM\n", i_split + 1, gguf_get_n_tensors(ctx_out), total_size);
i_split++;
}
}