Support input float tensor directly #1910
ggerganov
merged
20 commits into
ggml-org:master
from
ningshanwutuobang:embd_inp
add interface for float input
50ce2966
fixed inpL shape and type
5673a8de
add examples of input floats
20d5eef8
add test example for embd input
a9148709
fixed sampling
9c6117cd
add free for context
ba1f617d
fixed add end condition for generating
6ed48933
add examples for llava.py
8cea3ab9
add READMD for llava.py
4f1aa3cc
add READMD for llava.py
93c57a05
Merge remote-tracking branch 'origin/master' into embd_inp
46490c7a
add example of PandaGPT
53dfbbf5
refactor the interface and fixed the styles
9d866118
Merge branch 'master' into embd_inp
9b03f859
add cmake build for embd-input
beca5a61
add cmake build for embd-input
6cb62ca2
Add MiniGPT-4 example
40340d82
ggerganov
commented
on 2023-06-26
ggerganov
approved these changes
on 2023-06-26
change the order of the args of llama_eval_internal
39011ad7
Merge remote-tracking branch 'origin/master' into embd_inp
d99d3998
fix ci error
7abb513e
Login to write a write a comment.
Login via GitHub