Xin Qiu
58208a5883
Update FAQ document. ( #10300 )
...
* Update install_gpu.md
* Update resolve_error.md
* Update README.md
* Update resolve_error.md
* Update README.md
* Update resolve_error.md
2024-03-04 08:35:11 +08:00
Yuwen Hu
27d9a14989
[LLM] all-on-one update: memory optimize and streaming output ( #10302 )
...
* Memory saving for continous in-out pair run and add support for streaming output on MTL iGPU
* Small fix
* Small fix
* Add things back
2024-03-01 18:02:30 +08:00
SONG Ge
0ab40917fb
[LLM] Split merged_qk to separated q/k linear ( #10299 )
...
* modify merge_qk_linear to separated q/k linear
* update
2024-03-01 16:48:55 +08:00
Yang Wang
f4d7dbcde2
use fused qkv forward in qwen2 ( #10185 )
...
* use fused qkv forward in qwen2
* support both
* fix style
* fix rope
* remove pring
* fix style
* clean up
2024-03-01 16:46:35 +08:00
Xin Qiu
509e206de0
update doc about gemma random and unreadable output. ( #10297 )
...
* Update install_gpu.md
* Update README.md
* Update README.md
2024-03-01 15:41:16 +08:00
Wang, Jian4
beb9433cec
LLM: Reduce speculative _ipex_optimize_model memory use ( #10281 )
...
* use tpp
* update ipex
2024-03-01 13:48:23 +08:00
Yuwen Hu
f0ff0eebe1
[LLM] Support quantize kv cache for Baichuan2 7B ( #10280 )
...
* Add quatized kv cache framework for Baichuan2 7B
* Support quantize kv cache for baichuan2
* Small fix
* Fix python style
2024-03-01 13:35:42 +08:00
SONG Ge
273de341d7
hot-fix silu error import ( #10292 )
2024-03-01 10:11:37 +08:00
Shengsheng Huang
bcfad555df
revise llamaindex readme ( #10283 )
2024-02-29 17:19:23 +08:00
Xin Qiu
232273a1b5
Enable Gemma fused mlp + Gelu ( #10276 )
...
* update llama mlp forward
* add all
* fix style check
* split
* update
* update
* update
* fix style
2024-02-29 16:53:24 +08:00
Guancheng Fu
2d930bdca8
Add vLLM bf16 support ( #10278 )
...
* add argument load_in_low_bit
* add docs
* modify gpu doc
* done
---------
Co-authored-by: ivy-lv11 <lvzc@lamda.nju.edu.cn>
2024-02-29 16:33:42 +08:00
SONG Ge
13b0bc9075
[LLM] Add quantize_kv optimization for yuan2 model ( #10243 )
...
* add initial quantize_kv support for yuan2 model
* fix yuan2 quantize_kv generation
* apply fp16 conv layer optimizations
* disable mlp for quantize_kv
2024-02-29 16:33:26 +08:00
Zhicun
4e6cc424f1
Add LlamaIndex RAG ( #10263 )
...
* run demo
* format code
* add llamaindex
* add custom LLM with bigdl
* update
* add readme
* begin ut
* add unit test
* add license
* add license
* revised
* update
* modify docs
* remove data folder
* update
* modify prompt
* fixed
* fixed
* fixed
2024-02-29 15:21:19 +08:00
Jin Qiao
5d7243067c
LLM: add Baichuan2-13B-Chat 2048-256 to MTL perf ( #10273 )
2024-02-29 13:48:55 +08:00
Ruonan Wang
a9fd20b6ba
LLM: Update qkv fusion for GGUF-IQ2 ( #10271 )
...
* first commit
* update mistral
* fix transformers==4.36.0
* fix
* disable qk for mixtral now
* fix style
2024-02-29 12:49:53 +08:00
Jiao Wang
6fb65bb9d2
fix in transformers 4.36 ( #10150 )
2024-02-28 18:43:01 -08:00
Shengsheng Huang
43dac97e03
Update README.md ( #10260 )
2024-02-29 10:41:14 +08:00
Ruonan Wang
4b08bc1417
LLM: relax batch check of flash atttention by double check attention mask ( #10270 )
...
* relax batch check
* fix
* fix style
2024-02-29 09:39:55 +08:00
Yina Chen
07f36fbfcc
Fix gptj failed to extend ( #10269 )
2024-02-29 09:39:27 +08:00
Yishuo Wang
cccb02dad1
fix baichuan2 13b 2k input ( #10267 )
2024-02-28 17:20:20 +08:00
Heyang Sun
7244fd1ba5
Fix Arc StarCoder wrong query_shape when input is long ( #10268 )
...
* Fix Arc StarCoder wrong query_shape when input is long
* Update gptbigcode.py
2024-02-28 17:07:08 +08:00
Cengguang Zhang
a4de3095f3
LLM: Support quantize kv cache in mistral. ( #10261 )
...
* init
* update quantize kv.
2024-02-28 14:08:08 +08:00
Shengsheng Huang
db0d129226
Revert "Add rwkv example ( #9432 )" ( #10264 )
...
This reverts commit 6930422b42
2024-02-28 11:48:31 +08:00
Yining Wang
6930422b42
Add rwkv example ( #9432 )
...
* codeshell fix wrong urls
* restart runner
* add RWKV CPU & GPU example (rwkv-4-world-7b)
* restart runner
* update submodule
* fix runner
* runner-test
---------
Co-authored-by: Shengsheng Huang <shengsheng.huang@intel.com>
2024-02-28 11:41:00 +08:00
Keyan (Kyrie) Zhang
59861f73e5
Add Deepseek-6.7B ( #9991 )
...
* Add new example Deepseek
* Add new example Deepseek
* Add new example Deepseek
* Add new example Deepseek
* Add new example Deepseek
* modify deepseek
* modify deepseek
* Add verified model in README
* Turn cpu_embedding=True in Deepseek example
---------
Co-authored-by: Shengsheng Huang <shengsheng.huang@intel.com>
2024-02-28 11:36:39 +08:00
Yuxuan Xia
2524273198
Update AutoGen README ( #10255 )
...
* Update AutoGen README
* Fix AutoGen README typos
* Update AutoGen README
* Update AutoGen README
2024-02-28 11:34:45 +08:00
Zheng, Yi
2347f611cf
Add cpu and gpu examples of Mamba ( #9797 )
...
* Add mamba cpu example
* Add mamba gpu example
* Use a smaller model as the example
* minor fixes
---------
Co-authored-by: Shengsheng Huang <shengsheng.huang@intel.com>
2024-02-28 11:33:29 +08:00
Zhao Changmin
937e1f7c74
rebase ( #9104 )
...
Co-authored-by: leonardozcm <leonardozcm@gmail.com>
2024-02-28 11:18:21 +08:00
JunX
4833067489
fix GPU example link in README.md ( #9533 )
...
* fix GPU example link in README.md
* fix GPU links in llm README.md
2024-02-28 11:13:18 +08:00
Zhicun
308e637d0d
Add DeepSeek-MoE-16B-Chat ( #10155 )
...
* dsmoe-hf add
* add dsmoe pytorch
* update README
* modify comment
* remove GPU example
* update model name
* format code
2024-02-28 10:12:09 +08:00
Guoqiong Song
f4a2e32106
Stream llm example for both GPU and CPU ( #9390 )
2024-02-27 15:54:47 -08:00
Yang Wang
c581c6db30
draft mmint4 ( #10031 )
...
change to llm.cpp
support transposed format
revert
implement qkv fuse
fix style
change to vertically pack
change to enable_xetla
fix mlp_fusion_check
remove comments
address comments
add some comments
fix style
2024-02-27 14:55:16 -08:00
hxsz1997
cba61a2909
Add html report of ppl ( #10218 )
...
* remove include and language option, select the corresponding dataset based on the model name in Run
* change the nightly test time
* change the nightly test time of harness and ppl
* save the ppl result to json file
* generate csv file and print table result
* generate html
* modify the way to get parent folder
* update html in parent folder
* add llm-ppl-summary and llm-ppl-summary-html
* modify echo single result
* remove download fp16.csv
* change model name of PR
* move ppl nightly related files to llm/test folder
* reformat
* seperate make_table from make_table_and_csv.py
* separate make_csv from make_table_and_csv.py
* update llm-ppl-html
* remove comment
* add Download fp16.results
2024-02-27 17:37:08 +08:00
Zhicun
6d60982746
Env script: add license ( #10257 )
...
* env script
* update README.md
* modify README
* modify cpu info output
* add env-check.sh
* add env-check.bat
* add windows
* modify bat
* add license
2024-02-27 15:29:20 +08:00
Yishuo Wang
b4fa4ab46f
optimize yuan 2.0 again ( #10252 )
2024-02-27 14:51:42 +08:00
Zhicun
03b9c4930a
UX: Script to print env info ( #10088 )
...
* env script
* update README.md
* modify README
* modify cpu info output
* add env-check.sh
* add env-check.bat
* add windows
* modify bat
2024-02-27 14:45:36 +08:00
Keyan (Kyrie) Zhang
843fe546b0
Add CPU and GPU examples for DeciLM-7B ( #9867 )
...
* Add cpu and gpu examples for DeciLM-7B
* Add cpu and gpu examples for DeciLM-7B
* Add DeciLM-7B to README table
* modify deciLM
* modify deciLM
* modify deciLM
* Add verified model in README
* Add cpu_embedding=True
2024-02-27 13:15:49 +08:00
Yuwen Hu
38ae4b372f
Add yuan2-2b to win igpu perf test ( #10250 )
2024-02-27 11:08:33 +08:00
Heyang Sun
36a9e88104
Speculative Starcoder on CPU ( #10138 )
...
* Speculative Starcoder on CPU
* enable kv-cache pre-allocation
* refine codes
* refine
* fix style
* fix style
* fix style
* refine
* refine
* Update speculative.py
* Update gptbigcode.py
* fix style
* Update speculative.py
* enable mixed-datatype layernorm on top of torch API
* adaptive dtype
* Update README.md
2024-02-27 09:57:29 +08:00
Yishuo Wang
a47989c860
optimize yuan 2.0 performance ( #10244 )
2024-02-26 17:20:10 +08:00
Wang, Jian4
6c74b99a28
LLM: Update qwen readme ( #10245 )
2024-02-26 17:03:09 +08:00
hxsz1997
15ad2fd72e
Merge pull request #10226 from zhentaocc/fix_harness
...
Fix harness
2024-02-26 16:49:27 +08:00
Wang, Jian4
f9b75f900b
LLM: Enable qwen target_model ipex ( #10232 )
...
* change order
* enable qwen ipex
* update qwen example
* update
* fix style
* update
2024-02-26 16:41:12 +08:00
Jin Qiao
3e6d188553
LLM: add baichuan2-13b to mtl perf ( #10238 )
2024-02-26 15:55:56 +08:00
Yuwen Hu
e38e29511c
[LLM] Yuan2 MLP and Rotary optimization ( #10231 )
...
* Add optimization for rotary embedding
* Add mlp fused optimizatgion
* Python style fix
* Fix rotary embedding due to logits difference
* Small fix
2024-02-26 15:10:08 +08:00
Ziteng Zhang
ea23afc8ec
[LLM]update ipex part in mistral example readme ( #10239 )
...
* update ipex part in mistral example readme
2024-02-26 14:35:20 +08:00
SONG Ge
df2f3885ba
[LLM] Enable kv_cache and forward_qkv optimizations for yuan2 ( #10225 )
...
* add init kv_cache support for yuan2
* add forward qkv in yuan
2024-02-26 11:29:48 +08:00
Xiangyu Tian
85a99e13e8
LLM: Fix ChatGLM3 Speculative Example ( #10236 )
...
Fix ChatGLM3 Speculative Example.
2024-02-26 10:57:28 +08:00
Chen, Zhentao
213ef06691
fix readme
2024-02-24 00:38:08 +08:00
Ruonan Wang
28513f3978
LLM: support fp16 embedding & add mlp fusion for iq2_xxs ( #10219 )
...
* add fp16 embed
* small fixes
* fix style
* fix style
* fix comment
2024-02-23 17:26:24 +08:00
Yuwen Hu
eeecd9fc08
Python style fix ( #10230 )
2024-02-23 17:21:23 +08:00
Yuwen Hu
e511bbd8f1
[LLM] Add basic optimization framework for Yuan2 ( #10227 )
...
* Add basic optimization framework for Yuan2
* Small fix
* Python style fix
* Small fix
* Small fix
2024-02-23 17:05:00 +08:00
Xin Qiu
8ef5482da2
update Gemma readme ( #10229 )
...
* Update README.md
* Update README.md
* Update README.md
* Update README.md
2024-02-23 16:57:08 +08:00
Chen, Zhentao
6fe5344fa6
separate make_csv from the file
2024-02-23 16:33:38 +08:00
Chen, Zhentao
bfa98666a6
fall back to make_table.py
2024-02-23 16:33:38 +08:00
Ruonan Wang
19260492c7
LLM: fix action/installation error of mpmath ( #10223 )
...
* fix
* test
* fix
* update
2024-02-23 16:14:53 +08:00
Xin Qiu
aabfc06977
add gemma example ( #10224 )
...
* add gemma gpu example
* Update README.md
* add cpu example
* Update README.md
* Update README.md
* Update generate.py
* Update generate.py
2024-02-23 15:20:57 +08:00
yb-peng
a2c1675546
Add CPU and GPU examples for Yuan2-2B-hf ( #9946 )
...
* Add a new CPU example of Yuan2-2B-hf
* Add a new CPU generate.py of Yuan2-2B-hf example
* Add a new GPU example of Yuan2-2B-hf
* Add Yuan2 to README table
* In CPU example:1.Use English as default prompt; 2.Provide modified files in yuan2-2B-instruct
* In GPU example:1.Use English as default prompt;2.Provide modified files
* GPU example:update README
* update Yuan2-2B-hf in README table
* Add CPU example for Yuan2-2B in Pytorch-Models
* Add GPU example for Yuan2-2B in Pytorch-Models
* Add license in generate.py; Modify README
* In GPU Add license in generate.py; Modify README
* In CPU yuan2 modify README
* In GPU yuan2 modify README
* In CPU yuan2 modify README
* In GPU example, updated the readme for Windows GPU supports
* In GPU torch example, updated the readme for Windows GPU supports
* GPU hf example README modified
* GPU example README modified
2024-02-23 14:09:30 +08:00
yb-peng
f1f4094a09
Add CPU and GPU examples of phi-2 ( #10014 )
...
* Add CPU and GPU examples of phi-2
* In GPU hf example, updated the readme for Windows GPU supports
* In GPU torch example, updated the readme for Windows GPU supports
* update the table in BigDL/README.md
* update the table in BigDL/python/llm/README.md
2024-02-23 14:05:53 +08:00
Chen, Zhentao
f315c7f93a
Move harness nightly related files to llm/test folder ( #10209 )
...
* move harness nightly files to test folder
* change workflow file path accordingly
* use arc01 when pr
* fix path
* fix fp16 csv path
2024-02-23 11:12:36 +08:00
Xin Qiu
30795bdfbc
Gemma optimization: rms_norm, kv_cache, fused_rope, fused_rope+qkv ( #10212 )
...
* gemma optimization
* update
* update
* fix style
* meet code review
2024-02-23 10:07:24 +08:00
Guoqiong Song
63681af97e
falcon for transformers 4.36 ( #9960 )
...
* falcon for transformers 4.36
2024-02-22 17:04:40 -08:00
Jason Dai
84d5f40936
Update README.md ( #10213 )
2024-02-22 17:22:59 +08:00
Yina Chen
ce5840a8b7
GPT-J rope optimization on xpu ( #10182 )
...
* optimize
* update
* fix style & move use_fuse_rope
* add ipex version check
* fix style
* update
* fix style
* meet comments
* address comments
* fix style
2024-02-22 16:25:12 +08:00
Xiangyu Tian
f445217d02
LLM: Update IPEX to 2.2.0+cpu and Refactor for _ipex_optimize ( #10189 )
...
Update IPEX to 2.2.0+cpu and refactor for _ipex_optimize.
2024-02-22 16:01:11 +08:00
Heyang Sun
c876d9b5ca
Support for MPT rotary embedding ( #10208 )
2024-02-22 15:16:31 +08:00
Ruonan Wang
5e1fee5e05
LLM: add GGUF-IQ2 examples ( #10207 )
...
* add iq2 examples
* small fix
* meet code review
* fix
* meet review
* small fix
2024-02-22 14:18:45 +08:00
Yuwen Hu
21de2613ce
[LLM] Add model loading time record for all-in-one benchmark ( #10201 )
...
* Add model loading time record in csv for all-in-one benchmark
* Small fix
* Small fix to number after .
2024-02-22 13:57:18 +08:00
Ovo233
60e11b6739
LLM: Add mlp layer unit tests ( #10200 )
...
* add mlp layer unit tests
* add download baichuan-13b
* exclude llama for now
* install additional packages
* rename bash file
* switch to Baichuan2
* delete attention related code
* fix name errors in yml file
2024-02-22 13:44:45 +08:00
SONG Ge
ca1166a0e5
[LLM] Add quantize kv_cache for Baichuan2-13B ( #10203 )
...
* add quantize kv_cache for baichuan2-13b
* style fix
2024-02-22 13:43:35 +08:00
Ruonan Wang
34ee1aa91f
LLM: add esimd sdp support for chatglm3 ( #10205 )
...
* add esimd sdp support
* fix style
2024-02-22 13:37:16 +08:00
Yuxuan Xia
7cbc2429a6
Fix C-Eval ChatGLM loading issue ( #10206 )
...
* Add c-eval workflow and modify running files
* Modify the chatglm evaluator file
* Modify the ceval workflow for triggering test
* Modify the ceval workflow file
* Modify the ceval workflow file
* Modify ceval workflow
* Adjust the ceval dataset download
* Add ceval workflow dependencies
* Modify ceval workflow dataset download
* Add ceval test dependencies
* Add ceval test dependencies
* Correct the result print
* Fix the nightly test trigger time
* Fix ChatGLM loading issue
2024-02-22 10:00:43 +08:00
Yuwen Hu
94cb16fe40
[LLM] Small updates to Win GPU Install Doc ( #10199 )
...
* Make Offline installer as default for win gpu doc for oneAPI
* Small other fixes
2024-02-21 17:58:40 +08:00
binbin Deng
9975b029c5
LLM: add qlora finetuning example using trl.SFTTrainer ( #10183 )
2024-02-21 16:40:04 +08:00
Ruonan Wang
f7c96b19ef
LLM: support iq2 for mixtral ( #10191 )
...
* support name mapping for mixtral
* support mixtral mixed quantization
* fix style
* fix
2024-02-21 16:00:29 +08:00
yb-peng
b1a97b71a9
Harness eval: Add is_last parameter and fix logical operator in highlight_vals ( #10192 )
...
* Add is_last parameter and fix logical operator in highlight_vals
* Add script to update HTML files in parent folder
* Add running update_html_in_parent_folder.py in summarize step
* Add licence info
* Remove update_html_in_parent_folder.py in Summarize the results for pull request
2024-02-21 14:45:32 +08:00
Zhicun
c7e839e66c
Add Qwen1.5-7B-Chat ( #10113 )
...
* add Qwen1.5-7B-Chat
* modify Qwen1.5 example
* update README
* update prompt format
* update folder name and example README
* add Chinese prompt sample output
* update link in README
* correct the link
* update transformer version
2024-02-21 13:29:29 +08:00
Xin Qiu
56ad781f2f
qwen2 cpu fix ( #10187 )
2024-02-21 11:23:51 +08:00
Chen, Zhentao
39d37bd042
upgrade harness package version in workflow ( #10188 )
...
* upgrade harness
* update readme
2024-02-21 11:21:30 +08:00
Yuwen Hu
001c13243e
[LLM] Add support for low_low_bit benchmark on Windows GPU ( #10167 )
...
* Add support for low_low_bit performance test on Windows GPU
* Small fix
* Small fix
* Save memory during converting model process
* Drop the results for first time when loading in low bit on mtl igpu for better performance
* Small fix
2024-02-21 10:51:52 +08:00
Ziteng Zhang
276ef0e885
Speculative Ziya on CPU ( #10160 )
...
* Speculative Ziya on CPU
* Without part of Accelerate with BIGDL_OPT_IPEX
2024-02-21 10:30:39 +08:00
Zhao Changmin
4fbf449c2d
for rwkv4 ( #10179 )
2024-02-21 10:11:10 +08:00
yb-peng
de3dc609ee
Modify harness evaluation workflow ( #10174 )
...
* Modify table head in harness
* Specify the file path of fp16.csv
* change run to run nightly and run pr to debug
* Modify the way to get fp16.csv to downloading from github
* Change the method to calculate diff in html table
* Change the method to calculate diff in html table
* Re-arrange job order
* Re-arrange job order
* Change limit
* Change fp16.csv path
* Change highlight rules
* Change limit
2024-02-20 18:55:43 +08:00
Ruonan Wang
3288acb8de
LLM : Support embedding quantization (only q2k now) ( #10170 )
...
* basic logic added
* basic support
* support save&load, update mixed strategy
* fix style
* use int8 for lm_head
* add check for xpu
2024-02-20 16:56:57 +08:00
hxsz1997
6e10d98a8d
Fix some typos ( #10175 )
...
* add llm-ppl workflow
* update the DATASET_DIR
* test multiple precisions
* modify nightly test
* match the updated ppl code
* add matrix.include
* fix the include error
* update the include
* add more model
* update the precision of include
* update nightly time and add more models
* fix the workflow_dispatch description, change default model of pr and modify the env
* modify workflow_dispatch language options
* modify options
* modify language options
* modeify workflow_dispatch type
* modify type
* modify the type of language
* change seq_len type
* fix some typos
* revert changes to stress_test.txt
2024-02-20 14:14:53 +08:00
Zhicun
add3899311
Add ziya CPU example ( #10114 )
...
* ziya on CPU
* add README for ziya
* specify use_cache
* add arc CPU
* update prompt format
* update link
* add comments to emphasize use_cache
* update pip cmd
2024-02-20 13:59:52 +08:00
binbin Deng
2bb96c775c
LLM: fix device setting during saving optimized model ( #10154 )
2024-02-20 09:52:59 +08:00
Xin Qiu
1f6d5b9f30
enable fused rmsnorm and rope qwen2 ( #10163 )
...
* qwen2
* change convert
* cleanup
2024-02-20 08:33:09 +08:00
yb-peng
e31210ba00
Modify html table style and add fp16.csv in harness ( #10169 )
...
* Specify the version of pandas in harness evaluation workflow
* Specify the version of pandas in harness evaluation workflow
* Modify html table style and add fp16.csv in harness
* Modify comments
2024-02-19 18:13:40 +08:00
WeiguangHan
6c09aed90d
LLM: add qwen_1.5_7b model for arc perf test ( #10166 )
...
* LLM: add qwen_1.5_7b model for arc perf test
* small fix
* revert some codes
2024-02-19 17:21:00 +08:00
Yuxuan Xia
209122559a
Add Ceval workflow and modify the result printing ( #10140 )
...
* Add c-eval workflow and modify running files
* Modify the chatglm evaluator file
* Modify the ceval workflow for triggering test
* Modify the ceval workflow file
* Modify the ceval workflow file
* Modify ceval workflow
* Adjust the ceval dataset download
* Add ceval workflow dependencies
* Modify ceval workflow dataset download
* Add ceval test dependencies
* Add ceval test dependencies
* Correct the result print
2024-02-19 17:06:53 +08:00
Zhao Changmin
f8730e8dc1
Skip rescale rwkv linear when load_low_bit ( #10164 )
...
* rwkv_ld
2024-02-19 15:56:42 +08:00
Heyang Sun
3e2af5ec0a
Fix IPEX Baichuan Speculative ( #10162 )
...
* Fix IPEX Baichuan Speculative
* compatible with 13B
* Update speculative.py
2024-02-19 15:27:34 +08:00
Yina Chen
23c91cdce6
[LLM] Add min_step_draft in speculative decoding ( #10142 )
...
* Fix gptj kvcache & position id
* Add min_draft_tokens in speculative decoding
* fix style
* update
2024-02-19 14:31:41 +08:00
Chen, Zhentao
14ba2c5135
Harness: remove deprecated files ( #10165 )
2024-02-19 14:27:49 +08:00
Wang, Jian4
d3591383d5
LLM : Add CPU chatglm3 speculative example ( #10004 )
...
* init chatglm
* update
* update
2024-02-19 13:38:52 +08:00
Wang, Jian4
f2417e083c
LLM: enable chatglm3-6b target_model ipex ( #10085 )
...
* init
* always make casual_mask
* not return last tensor
* update
* optimize_model = False
* enable optimized=False
* enable optimized_model=true
* speed_up ipex target_model
* remove if True
* use group_size
* update python style
* update
* update
2024-02-19 13:38:32 +08:00
Heyang Sun
177273c1a4
IPEX Speculative Support for Baichuan2 7B ( #10112 )
...
* IPEX Speculative Support for Baichuan2 7B
* fix license problems
* refine
2024-02-19 09:12:57 +08:00
Yina Chen
1508d6b089
Fix gptj kvcache & position id ( #10141 )
2024-02-18 10:02:49 +08:00
yb-peng
b4dc33def6
In harness-evaluation workflow, add statistical tables ( #10118 )
...
* chnage storage
* fix typo
* change label
* change label to arc03
* change needs in the last step
* add generate csv in harness/make_table_results.py
* modify needs in the last job
* add csv to html
* mfix path issue in llm-harness-summary-nightly
* modify output_path
* modify args in make_table_results.py
* modify make table command in summary
* change pr env label
* remove irrelevant code in summary; add set output path step; add limit in harness run
* re-organize code structure
* modify limit in run harness
* modify csv_to_html input path
* modify needs in summary-nightly
2024-02-08 19:01:05 +08:00
Yishuo Wang
4d33aac7f9
quick fix qwen2 fp8 kv cache ( #10135 )
2024-02-08 17:04:59 +08:00
Cengguang Zhang
39d90839aa
LLM: add quantize kv cache for llama. ( #10086 )
...
* feat: add quantize kv cache for llama.
* fix style.
* add quantized attention forward function.
* revert style.
* fix style.
* fix style.
* update quantized kv cache and add quantize_qkv
* fix style.
* fix style.
* optimize quantize kv cache.
* fix style.
2024-02-08 16:49:22 +08:00
Yishuo Wang
d848efe17c
add quantize kv cache support for qwen2 ( #10134 )
2024-02-08 16:17:21 +08:00
SONG Ge
3f79128ed7
[LLM] Enable kv_cache optimization for Qwen2 on transformers-v4.37.0 ( #10131 )
...
* add support for kv_cache optimization on transformers-v4.37.0
* enable attention forward
* style fix
* disable rotary for now
2024-02-08 14:20:26 +08:00
Ruonan Wang
063dc145ac
LLM: basic support for q2k ( #10132 )
...
* basic support for q2k
* fix style
2024-02-08 13:52:01 +08:00
binbin Deng
11fe5a87ec
LLM: add Modelscope model example ( #10126 )
2024-02-08 11:18:07 +08:00
Cengguang Zhang
0cf6a12691
LLM: add default torch_dtype for fp16. ( #10124 )
...
* set default torch_dtype for fp16.
* fix style.
* bug fix.
* update bug fix.
2024-02-08 10:24:16 +08:00
Yishuo Wang
1aa0c623ce
disable fused layer norm on UHD ( #10130 )
2024-02-08 10:20:01 +08:00
Yuwen Hu
a8450fc300
[LLM] Support MLP optimization for Qwen1.5 ( #10123 )
2024-02-08 09:15:34 +08:00
Yuwen Hu
81ed65fbe7
[LLM] Add qwen1.5-7B in iGPU perf ( #10127 )
...
* Add qwen1.5 test config yaml with transformers 4.37.0
* Update for yaml file
2024-02-07 22:31:20 +08:00
Jin Qiao
0fcfbfaf6f
LLM: add rwkv5 eagle GPU HF example ( #10122 )
...
* LLM: add rwkv5 eagle example
* fix
* fix link
2024-02-07 16:58:29 +08:00
binbin Deng
925f82107e
LLM: support models hosted by modelscope ( #10106 )
2024-02-07 16:46:36 +08:00
binbin Deng
c1ec3d8921
LLM: update FAQ about too many open files ( #10119 )
2024-02-07 15:02:24 +08:00
Keyan (Kyrie) Zhang
2e80701f58
Unit test on final logits and the logits of the last attention layer ( #10093 )
...
* Add unit test on final logits and attention
* Add unit test on final logits and attention
* Modify unit test on final logits and attention
2024-02-07 14:25:36 +08:00
Yuxuan Xia
3832eb0ce0
Add ChatGLM C-Eval Evaluator ( #10095 )
...
* Add ChatGLM ceval evaluator
* Modify ChatGLM Evaluator Reference
2024-02-07 11:27:06 +08:00
Jin Qiao
63050c954d
fix ( #10117 )
2024-02-07 11:05:11 +08:00
Jin Qiao
d3d2ee1b63
LLM: add speech T5 GPU example ( #10090 )
...
* add speech t5 example
* fix
* fix
2024-02-07 10:50:02 +08:00
Jin Qiao
2f4c754759
LLM: add bark gpu example ( #10091 )
...
* add bark gpu example
* fix
* fix license
* add bark
* add example
* fix
* another way
2024-02-07 10:47:11 +08:00
Xiangyu Tian
8953acd7d6
[LLM] Fix log condition for BIGDL_OPT_IPEX ( #10115 )
...
Fix log condition for BIGDL_OPT_IPEX
2024-02-07 10:27:10 +08:00
SONG Ge
0eccb94d75
remove text-generation-webui from bigdl repo ( #10107 )
2024-02-06 17:46:52 +08:00
Ovo233
2aaa21c41d
LLM: Update ppl tests ( #10092 )
...
* update ppl tests
* use load_dataset api
* add exception handling
* add language argument
* address comments
2024-02-06 17:31:48 +08:00
Yuwen Hu
3a46b57253
[LLM] Add RWKV4 HF GPU Example ( #10105 )
...
* Add GPU HF example for RWKV 4
* Add link to rwkv4
* fix
2024-02-06 16:30:24 +08:00
Yuwen Hu
518ef95abc
Small fix for Nonetype error ( #10104 )
2024-02-06 14:58:52 +08:00
Ruonan Wang
d61f4905ac
LLM: 2bit quantization initial support ( #10042 )
...
* basis quantize support
* fix new module name
* small update
* and mixed int4 with iq2_xxs
* remove print
* code refactor
* fix style
* meet code review
2024-02-06 14:58:32 +08:00
dingbaorong
36c9442c6d
Arc Stable version test ( #10087 )
...
* add batch_size in stable version test
* add batch_size in excludes
* add excludes for batch_size
* fix ci
* triger regression test
* fix xpu version
* disable ci
* address kai's comment
---------
Co-authored-by: Ariadne <wyn2000330@126.com>
2024-02-06 10:23:50 +08:00
Jiao Wang
33b9e7744d
fix dimension ( #10097 )
2024-02-05 15:07:38 -08:00
SONG Ge
4b02ff188b
[WebUI] Add prompt format and stopping words for Qwen ( #10066 )
...
* add prompt format and stopping_words for qwen mdoel
* performance optimization
* optimize
* update
* meet comments
2024-02-05 18:23:13 +08:00
WeiguangHan
0aecd8637b
LLM: small fix for the html script ( #10094 )
2024-02-05 17:27:34 +08:00
Zhicun
7d2be7994f
add phixtral and optimize phi-moe ( #10052 )
2024-02-05 11:12:47 +08:00
Zhicun
676d6923f2
LLM: modify transformersembeddings.embed() in langchain ( #10051 )
2024-02-05 10:42:10 +08:00
Jin Qiao
ad050107b3
LLM: fix mpt load_low_bit issue ( #10075 )
...
* fix
* retry
* retry
2024-02-05 10:17:07 +08:00
SONG Ge
9050991e4e
fix gradio check issue temply ( #10082 )
2024-02-04 16:46:29 +08:00
WeiguangHan
c2e562d037
LLM: add batch_size to the csv and html ( #10080 )
...
* LLM: add batch_size to the csv and html
* small fix
2024-02-04 16:35:44 +08:00
binbin Deng
7e49fbc5dd
LLM: make finetuning examples more common for other models ( #10078 )
2024-02-04 16:03:52 +08:00
Heyang Sun
90f004b80b
remove benchmarkwrapper form deepspeed example ( #10079 )
2024-02-04 15:42:15 +08:00
Ruonan Wang
8e33cb0f38
LLM: support speecht5_tts ( #10077 )
...
* support speecht5_tts
* fix
2024-02-04 13:26:42 +08:00
ivy-lv11
428b7105f6
Add HF and PyTorch example InternLM2 ( #10061 )
2024-02-04 10:25:55 +08:00
Yina Chen
77be19bb97
LLM: Support gpt-j in speculative decoding ( #10067 )
...
* gptj
* support gptj in speculative decoding
* fix
* update readme
* small fix
2024-02-02 14:54:55 +08:00
SONG Ge
19183ef476
[WebUI] Reset bigdl-llm loader options with default value ( #10064 )
...
* reset bigdl-llm loader options with default value
* remove options which maybe complex for naive users
2024-02-01 15:45:39 +08:00
Xin Qiu
6e0f1a1e92
use apply_rotary_pos_emb_cache_freq_xpu in mixtral ( #10060 )
...
* use apply_rotary_pos_emb_cache_freq_xpu in mixtral
* fix style
2024-02-01 15:40:49 +08:00
binbin Deng
aae20d728e
LLM: Add initial DPO finetuning example ( #10021 )
2024-02-01 14:18:08 +08:00
Heyang Sun
601024f418
Mistral CPU example of speculative decoding ( #10024 )
...
* Mistral CPU example of speculative decoding
* update transformres version
* update example
* Update README.md
2024-02-01 10:52:32 +08:00
Heyang Sun
968e70544d
Enable IPEX Mistral in Speculative ( #10059 )
2024-02-01 10:48:16 +08:00
Yina Chen
3ca03d4e97
Add deepmind sample into bigdl-llm speculative decoding ( #10041 )
...
* migrate deepmind sample
* update
* meet comments
* fix style
* fix style
2024-02-01 09:57:02 +08:00
WeiguangHan
d2d3f6b091
LLM: ensure the result of daily arc perf test ( #10016 )
...
* ensure the result of daily arc perf test
* small fix
* small fix
* small fix
* small fix
* small fix
* small fix
* small fix
* small fix
* small fix
* small fix
* concat more csvs
* small fix
* revert some files
2024-01-31 18:26:21 +08:00
WeiguangHan
9724939499
temporarily disable bloom 2k input ( #10056 )
2024-01-31 17:49:12 +08:00
Jin Qiao
8c8fc148c9
LLM: add rwkv 5 ( #10048 )
2024-01-31 15:54:55 +08:00
WeiguangHan
a9018a0e95
LLM: modify the GPU example for redpajama model ( #10044 )
...
* LLM: modify the GPU example for redpajama model
* small fix
2024-01-31 14:32:08 +08:00
Yuxuan Xia
95636cad97
Add AutoGen CPU and XPU Example ( #9980 )
...
* Add AutoGen example
* Adjust AutoGen README
* Adjust AutoGen README
* Change AutoGen README
* Change AutoGen README
2024-01-31 11:31:18 +08:00
Heyang Sun
7284edd9b7
Vicuna CPU example of speculative decoding ( #10018 )
...
* Vicuna CPU example of speculative decoding
* Update speculative.py
* Update README.md
* add requirements for ipex
* Update README.md
* Update speculative.py
* Update speculative.py
2024-01-31 11:23:50 +08:00
Wang, Jian4
7e5cd42a5c
LLM : Update optimize ipex bf16 ( #10038 )
...
* use 4.35.2 and remove
* update rmsnorm
* remove
* remove
* update python style
* update
* update python style
* update
* fix style
* update
* remove whitespace
2024-01-31 10:59:55 +08:00
Wang, Jian4
fb53b994f8
LLM : Add llama ipex optimized ( #10046 )
...
* init ipex
* remove padding
2024-01-31 10:38:46 +08:00
Ruonan Wang
3685622f29
LLM: fix llama 4.36 forward( #10047 )
2024-01-31 10:31:10 +08:00
Yishuo Wang
53a5140eff
Optimize rwkv v5 rest token again ( #10043 )
2024-01-31 10:01:11 +08:00
Heyang Sun
b1ff28ceb6
LLama2 CPU example of speculative decoding ( #9962 )
...
* LLama2 example of speculative decoding
* add docs
* Update speculative.py
* Update README.md
* Update README.md
* Update speculative.py
* remove autocast
2024-01-31 09:45:20 +08:00
WeiguangHan
0fcad6ce14
LLM: add gpu example for redpajama models ( #10040 )
2024-01-30 19:39:28 +08:00
Xiangyu Tian
9978089796
[LLM] Enable BIGDL_OPT_IPEX in speculative baichuan2 13b example ( #10028 )
...
Enable BIGDL_OPT_IPEX in speculative baichuan2 13b example
2024-01-30 17:11:37 +08:00
Ovo233
226f398c2a
fix ppl test errors ( #10036 )
2024-01-30 16:26:21 +08:00
Xin Qiu
13e61738c5
hide detail memory for each token in benchmark_utils.py ( #10037 )
2024-01-30 16:04:17 +08:00
Ruonan Wang
6b63ba23d1
LLM: add full module name during convert ( #10035 )
2024-01-30 14:43:07 +08:00
Yishuo Wang
7dfa6dbe46
add rwkv time shift optimization ( #10032 )
2024-01-30 14:10:55 +08:00
Xiangyu Tian
f57d0fda8b
[LLM] Use IPEX Optimization for Self Speculative Decoding ( #9997 )
...
Use IPEX Optimization for Self Speculative Decoding
2024-01-30 09:11:06 +08:00
Ruonan Wang
ccf8f613fb
LLM: update fp16 Linear on ARC/FLEX ( #10023 )
2024-01-29 18:25:26 +08:00
Shaojun Liu
824c8029d7
Fix "local variable 'model' referenced before assignment" ( #10022 )
2024-01-29 16:18:04 +08:00
Heyang Sun
cc3f122f6a
Baichuan2 CPU example of speculative decoding ( #10003 )
...
* Baichuan2 CPU example of speculative decoding
* Update generate.py
* Update README.md
* Update generate.py
* Update generate.py
* Update generate.py
* fix default model
* fix wrong chinese coding
* Update generate.py
* update prompt
* update sample outputs
* baichuan 7b needs transformers==4.31.0
* rename example file's name
2024-01-29 14:21:09 +08:00
Xiangyu Tian
f37e4702bc
[LLM] Use IPEX Optimization for BF16 Model ( #9988 )
...
Use IPEX Optimization for BF16 Model by env BIGDL_OPT_IPEX=true
2024-01-29 11:28:25 +08:00
Jin Qiao
440cfe18ed
LLM: GPU Example Updates for Windows ( #9992 )
...
* modify aquila
* modify aquila2
* add baichuan
* modify baichuan2
* modify blue-lm
* modify chatglm3
* modify chinese-llama2
* modiy codellama
* modify distil-whisper
* modify dolly-v1
* modify dolly-v2
* modify falcon
* modify flan-t5
* modify gpt-j
* modify internlm
* modify llama2
* modify mistral
* modify mixtral
* modify mpt
* modify phi-1_5
* modify qwen
* modify qwen-vl
* modify replit
* modify solar
* modify starcoder
* modify vicuna
* modify voiceassistant
* modify whisper
* modify yi
* modify aquila2
* modify baichuan
* modify baichuan2
* modify blue-lm
* modify chatglm2
* modify chatglm3
* modify codellama
* modify distil-whisper
* modify dolly-v1
* modify dolly-v2
* modify flan-t5
* modify llama2
* modify llava
* modify mistral
* modify mixtral
* modify phi-1_5
* modify qwen-vl
* modify replit
* modify solar
* modify starcoder
* modify yi
* correct the comments
* remove cpu_embedding in code for whisper and distil-whisper
* remove comment
* remove cpu_embedding for voice assistant
* revert modify voice assistant
* modify for voice assistant
* add comment for voice assistant
* fix comments
* fix comments
2024-01-29 11:25:11 +08:00
Yuwen Hu
c6d4f91777
[LLM] Add UTs of load_low_bit for transformers-style API ( #10001 )
...
* Add uts for transformers api load_low_bit generation
* Small fixes
* Remove replit-code for CPU tests due to current load_low_bit issue on MPT
* Small change
* Small reorganization to llm unit tests on CPU
* Small fixes
2024-01-29 10:18:23 +08:00
Yishuo Wang
d720554d43
simplify quantize kv cache api ( #10011 )
2024-01-29 09:23:57 +08:00
Yina Chen
a3322e2a6c
add fp8 e5 to use_xmx ( #10015 )
2024-01-26 18:29:46 +08:00
Qiyuan Gong
9e18ea187f
[LLM] Avoid KV Cache OOM when seq len is larger than 1 ( #10006 )
...
* Avoid OOM during muti-round streaming chat with kv cache
* For llama like kv cache, i.e., [bs, n_head, seq_len, head_dim], use is_enough_kv_cache_room_4_31.
* Other models need to compare kv cache size with kv_len.
2024-01-26 17:30:08 +08:00
binbin Deng
e5ae6f2c13
LLM: fix truncation logic of past_key_values in chatglm multi turn chat ( #10007 )
...
* Avoid frequently truncating past_key_values when its length is larger than required.
2024-01-26 16:56:02 +08:00
Yuwen Hu
1eaaace2dc
Update perf test all-in-one config for batch_size arg ( #10012 )
2024-01-26 16:46:36 +08:00
Xin Qiu
7952bbc919
add conf batch_size to run_model ( #10010 )
2024-01-26 15:48:48 +08:00
SONG Ge
421e7cee80
[LLM] Add Text_Generation_WebUI Support ( #9884 )
...
* initially add text_generation_webui support
* add env requirements install
* add necessary dependencies
* update for starting webui
* update shared and noted to place models
* update heading of part3
* meet comments
* add copyright license
* remove extensions
* convert tutorial to windows side
* add warm-up to optimize performance
2024-01-26 15:12:49 +08:00
Yuwen Hu
f0da0c131b
Disable llama2 optimize model true or false test for now in Arc UTs ( #10008 )
2024-01-26 14:42:11 +08:00
Ruonan Wang
a00efa0564
LLM: add mlp & qkv fusion for FP16 Llama-7B ( #9932 )
...
* add mlp fusion for llama
* add mlp fusion
* fix style
* update
* add mm_qkv_out
* fix style
* update
* meet code review
* meet code review
2024-01-26 11:50:38 +08:00
Wang, Jian4
98ea3459e5
LLM : Fix llama draft_model dtype error ( #10005 )
...
* fix llama draft_model dtype error
* updat
2024-01-26 10:59:48 +08:00
Yishuo Wang
aae1870096
fix qwen kv cache length ( #9998 )
2024-01-26 10:15:01 +08:00
Chen, Zhentao
762adc4f9d
Reformat summary table ( #9942 )
...
* reformat the table
* refactor the file
* read result.json only
2024-01-25 23:49:00 +08:00
binbin Deng
171fb2d185
LLM: reorganize GPU finetuning examples ( #9952 )
2024-01-25 19:02:38 +08:00
Yishuo Wang
24b34b6e46
change xmx condition ( #10000 )
2024-01-25 17:48:11 +08:00
Ziteng Zhang
8b08ad408b
Add batch_size in all_in_one ( #9999 )
...
Add batch_size in all_in_one, except run_native_int4
2024-01-25 17:43:49 +08:00
Wang, Jian4
093e6f8f73
LLM: Add qwen CPU speculative example ( #9985 )
...
* init from gpu
* update for cpu
* update
* update
* fix xpu readme
* update
* update example prompt
* update prompt and add 72b
* update
* update
2024-01-25 17:01:34 +08:00
Yishuo Wang
bf65548d29
Add quantize kv cache support for chaglm2/3 ( #9996 )
2024-01-25 16:55:59 +08:00
Chen, Zhentao
86055d76d5
fix optimize_model not working ( #9995 )
2024-01-25 16:39:05 +08:00
Wang, Jian4
9bff84e6fd
LLM: Convert draft_model kv_cache from bf16 to fp32 ( #9964 )
...
* convert bf16 to fp32
* update
* change when init
* init first and cut off after
* init and exchange
* update python type
* update
* fix bug
* update
* update
2024-01-25 11:20:27 +08:00
Yina Chen
99ff6cf048
Update gpu spec decoding baichuan2 example dependency ( #9990 )
...
* add dependency
* update
* update
2024-01-25 11:05:04 +08:00
Yina Chen
27338540c3
Fix repetition_penalty not activated issue ( #9989 )
2024-01-25 10:40:41 +08:00
Jason Dai
3bc3d0bbcd
Update self-speculative readme ( #9986 )
2024-01-24 22:37:32 +08:00
Yuwen Hu
b27e5a27b9
Remove the check for meta device in _replace_with_low_bit_linear ( #9984 )
2024-01-24 18:15:39 +08:00
Ruonan Wang
d4f65a6033
LLM: add mistral speculative example ( #9976 )
...
* add mistral example
* update
2024-01-24 17:35:15 +08:00
Yina Chen
b176cad75a
LLM: Add baichuan2 gpu spec example ( #9973 )
...
* add baichuan2 gpu spec example
* update readme & example
* remove print
* fix typo
* meet comments
* revert
* update
2024-01-24 16:40:16 +08:00
Jinyi Wan
ec2d9de0ea
Fix README.md for solar ( #9957 )
2024-01-24 15:50:54 +08:00
Mingyu Wei
bc9cff51a8
LLM GPU Example Update for Windows Support ( #9902 )
...
* Update README in LLM GPU Examples
* Update reference of Intel GPU
* add cpu_embedding=True in comment
* small fixes
* update GPU/README.md and add explanation for cpu_embedding=True
* address comments
* fix small typos
* add backtick for cpu_embedding=True
* remove extra backtick in the doc
* add period mark
* update readme
2024-01-24 13:42:27 +08:00
Chen, Zhentao
e0db44dcb6
fix unexpected keyword argument 'device' ( #9982 )
...
* add device for chatglm3 only
* add comment for this change
* fix style
* fix style
* fix style again..
* finally fixed style
2024-01-24 13:20:46 +08:00
Mingyu Wei
50a851e3b3
LLM: separate arc ut for disable XMX ( #9953 )
...
* separate test_optimize_model api with disabled xmx
* delete test_optimize_model in test_transformers_api.py
* set env variable in .sh/ put back test_optimize_model
* unset env variable
* remove env setting in .py
* address errors in action
* remove import ipex
* lower tolerance
2024-01-23 19:04:47 +08:00
Yuwen Hu
8d28aa8e2b
[LLM] Fix the model.device problem when cpu_embedding=True ( #9971 )
...
* Overwrite the device attribute for CPUPinnedParam
* Expose cpu_embedding=True for Linux users
* Fix python style
2024-01-23 18:51:11 +08:00
Yishuo Wang
f82782cd3b
fix starcoder ( #9975 )
2024-01-23 17:24:53 +08:00
WeiguangHan
be5836bee1
LLM: fix outlier value ( #9945 )
...
* fix outlier value
* small fix
2024-01-23 17:04:13 +08:00
