github/gpt4all
1
0
Fork 0
mirror of https://github.com/nomic-ai/gpt4all.git synced 2025-06-25 06:53:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

remove unused llmodel_shared.h

Browse Source 
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
This commit is contained in:
Jared Van Bortel 2024-08-06 16:50:35 -04:00
parent de7cb36fcc
commit 43b6f63589
1 changed files with 0 additions and 49 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

49
gpt4all-backend/llmodel_shared.h
View File

@ -1,49 +0,0 @@
#pragma once
#include <ggml.h>
#include <cstddef>
#include <cstdint>
#include <vector>
struct llm_buffer {
uint8_t * addr = NULL;
size_t size = 0;
void resize(size_t size) {
delete[] addr;
addr = new uint8_t[size];
this->size = size;
}
~llm_buffer() {
delete[] addr;
}
};
struct llm_kv_cache {
struct ggml_tensor * k;
struct ggml_tensor * v;
struct ggml_context * ctx = NULL;
llm_buffer buf;
int n; // number of tokens currently in the cache
~llm_kv_cache() {
if (ctx) {
ggml_free(ctx);
}
}
};
inline void ggml_graph_compute_g4a(llm_buffer& buf, ggml_cgraph * graph, int n_threads)
{
struct ggml_cplan plan = ggml_graph_plan(graph, n_threads);
if (plan.work_size > 0) {
buf.resize(plan.work_size);
plan.work_data = buf.addr;
}
ggml_graph_compute(graph, &plan);
}