llama.cpp
ggml : change ggml_graph_compute() API to not require context
#1999
Merged
Merged
ggml : change ggml_graph_compute() API to not require context #1999
ggerganov
merged
20 commits into
ggml-org:master
from
mqy:ggml_graph_compute_context
mqy
commented
on 2023-07-03
mqy
commented
on 2023-07-03
mqy
changed the title
mqy
commented
on 2023-07-03
slaren
commented
on 2023-07-04
ggml_graph_compute: deprecate using ggml_context, try resolve issue #287
a1e7c692
rewrite: no longer consider backward compitability; plan and make_plan
b11ac01f
minor: rename ctx as plan; const
a37de239
remove ggml_graph_compute from tests/test-grad0.c, but current change…
db81f33e
add static ggml_graph_compute_sugar()
2b502c32
minor: update comments
cb1dec0e
reusable buffers
b1331d7e
ggml : more consistent naming + metal fixes
53cfb4b9
ggml : fix docs
4646cc2c
tests : disable grad / opt + minor naming changes
8e1f0b68
ggml : add ggml_graph_compute_with_ctx()
2392f7a9
ci : enable test-grad0
1b9994f8
examples : factor out plan allocation into a helper function
a67404e7
llama : factor out plan stuff into a helper function
2d3a5252
ci : fix env
8fdf86dd
llama : fix duplicate symbols + refactor example benchmark
9c9bdaf0
ggml : remove obsolete assert + refactor n_tasks section
8dc7f104
ggml : fix indentation in switch
551ed082
ggerganov
approved these changes
on 2023-07-06
llama : avoid unnecessary bool
f789f2ce
ggerganov
changed the title
ggml : remove comments from source file and match order in header
c15833c8
ggerganov
changed the title
Login to write a write a comment.
Login via GitHub