Commit graph

284 commits

Author SHA1 Message Date
qwopqwop200
f0f37c1fe7
fix bug 2023-05-01 18:09:39 +09:00
qwopqwop200
95e633a597
add old cuda 2023-05-01 13:05:14 +09:00
qwopqwop200
5a69e22a93
add qlinear_old 2023-05-01 13:04:47 +09:00
qwopqwop200
9dfcac8e26
add qlinear_old 2023-05-01 13:03:57 +09:00
潘其威(William)
5fa803334d
Merge branch 'main' into change-save-name 2023-04-29 20:36:45 +08:00
qwopqwop200
787909084f
fix bug 2023-04-29 19:08:34 +09:00
qwopqwop200
a2ef4b98db
change save the name 2023-04-29 18:20:46 +09:00
qwopqwop200
1792cd1111
change save the name 2023-04-29 18:16:48 +09:00
ZXED
24a371d14a
use the same Optional style as in other params 2023-04-29 09:52:11 +03:00
ZXED
c22770188d
allow user to set trust_remote_code flag manually 2023-04-29 09:52:11 +03:00
ZXED
b3f19a7ba7
support custom model name when loading the model 2023-04-29 09:52:11 +03:00
ZXED
ea8ab73343
support custom quantize_config when loading the model 2023-04-29 09:51:50 +03:00
PanQiWei
16d8dd200f remove non-parameters module from MOSSGPTQForCausalLM.outside_layer_modules 2023-04-29 10:58:29 +08:00
PanQiWei
b490ab004e remove override of _resize_attention_mask for llama and opt 2023-04-28 23:08:42 +08:00
qwopqwop200
ae8b1a22a3
change global to local 2023-04-28 23:18:39 +09:00
qwopqwop200
e914b9b1bd
update support 256 not div 2023-04-28 22:48:23 +09:00
qwopqwop200
c9215a1b5b
change div num 2023-04-28 22:42:29 +09:00
qwopqwop200
19f167e58b
add raise-exception 2023-04-28 22:24:44 +09:00
潘其威(William)
1e353a8dc5
Merge pull request #24 from PanQiWei/speedup_quantization
Offloading and Multiple devices quantization/inference
2023-04-28 18:50:12 +08:00
PanQiWei
bdb713b5a3 add batch_size to model.quant() api 2023-04-28 18:26:07 +08:00
PanQiWei
41564a48db make data_utils.py as global utils 2023-04-28 18:08:58 +08:00
PanQiWei
3dfc87bec3 return module in .to function 2023-04-28 17:20:46 +08:00
PanQiWei
a69a73a22c fix device mismatch when directly using model to inference after quantization 2023-04-28 16:41:46 +08:00
qwopqwop200
329a64ed40
support conv1d,conv2d 2023-04-28 09:15:42 +09:00
qwopqwop200
bb9afe8b61
support conv1d,conv2d 2023-04-28 09:15:13 +09:00
qwopqwop200
c1b7c7647d
support conv1d 2023-04-28 09:14:44 +09:00
qwopqwop200
ac41f68532
add gpt2 2023-04-28 09:14:05 +09:00
qwopqwop200
dad249990c
add gpt2 2023-04-28 09:13:22 +09:00
qwopqwop200
435eebee4b
support conv1d,conv2d 2023-04-28 09:13:00 +09:00
qwopqwop200
cc0f71a568
add gpt2 2023-04-28 09:11:50 +09:00
qwopqwop200
3f90a22632
fix bug 2023-04-28 08:26:58 +09:00
qwopqwop200
9c38393e31
fix bug about wf meta device 2023-04-28 08:26:11 +09:00
PanQiWei
d0cd5af5d3 make code more robust 2023-04-28 01:29:12 +08:00
PanQiWei
51d2e53130 add support to cpu offloading and multi gpus inference on quantized model 2023-04-28 00:53:57 +08:00
PanQiWei
b14dca9207 disk offload assertion 2023-04-27 21:31:53 +08:00
PanQiWei
7a3397e7ba add cpu offload when doing quantization 2023-04-27 21:25:24 +08:00
PanQiWei
ac3f7054e0 big fix 2023-04-27 19:33:25 +08:00
PanQiWei
498de923f2 support multi gpus quantization 2023-04-27 18:48:43 +08:00
qwopqwop200
8b6ee04aee
add option 2023-04-27 17:29:36 +09:00
PanQiWei
c9bb427546 align 'from_pretrained' api 2023-04-27 02:29:32 +08:00
PanQiWei
a2abff983e support dispatch layers to different devices when loading pretrained model before quantization 2023-04-27 02:24:08 +08:00
PanQiWei
950f203260 add 'n_positions' to sequence length search list 2023-04-27 01:09:10 +08:00
PanQiWei
893c3264cb make layer ignorance more robust 2023-04-26 19:35:19 +08:00
PanQiWei
f2359f56cb add support to use push_to_hub to upload and share quantized model 2023-04-26 16:55:01 +08:00
PanQiWei
bf2ae6768d bug fix 2023-04-26 13:33:56 +08:00
PanQiWei
73cb1dbf09 optimize import and format code 2023-04-26 13:08:47 +08:00
PanQiWei
975f100d0f init Quantizer() at GPTQ() init stage 2023-04-25 23:13:09 +08:00
PanQiWei
c35dce525e format code 2023-04-25 22:58:52 +08:00
PanQiWei
9f7f44146f format code 2023-04-25 22:45:27 +08:00
PanQiWei
b71211b4c3 format code 2023-04-25 22:36:28 +08:00