LLM: add baichuan2 example for arc (#8994)
* add baichuan2 examples * add link * small fix
This commit is contained in:
Ruonan Wang
GitHub
parent
7353882732
commit
cabe7c0358
4 changed files with 140 additions and 1 deletions
|
|
@ -5,6 +5,7 @@ You can use BigDL-LLM to run almost every Huggingface Transformer models with IN
|
|||
| Model | Example |
|
||||
|------------|----------------------------------------------------------|
|
||||
| Baichuan | [link](hf-transformers-models/baichuan) |
|
||||
| Baichuan2 | [link](hf-transformers-models/baichuan2) |
|
||||
| ChatGLM2 | [link](hf-transformers-models/chatglm2) |
|
||||
| Chinese Llama2 | [link](hf-transformers-models/chinese-llama2)|
|
||||
| Falcon | [link](hf-transformers-models/falcon) |
|
||||
|
|
|
|||
|
|
@ -4,7 +4,8 @@ You can use BigDL-LLM to run almost every Huggingface Transformer models with IN
|
|||
## Verified models
|
||||
| Model | Example |
|
||||
|------------|----------------------------------------------------------|
|
||||
| Baichuan | [link](baichuan) |
|
||||
| Baichuan | [link](baichuan) |
|
||||
| Baichuan2 | [link](baichuan2) |
|
||||
| ChatGLM2 | [link](chatglm2) |
|
||||
| Chinese Llama2 | [link](chinese-llama2)|
|
||||
| Falcon | [link](falcon) |
|
||||
|
|
|
|||
|
|
@ -0,0 +1,59 @@
|
|||
# Baichuan
|
||||
In this directory, you will find examples on how you could apply BigDL-LLM INT4 optimizations on Baichuan2 models on [Intel GPUs](../README.md). For illustration purposes, we utilize the [baichuan-inc/Baichuan2-7B-Chat](https://huggingface.co/baichuan-inc/Baichuan-7B-Chat) as a reference Baichuan model.
|
||||
|
||||
## 0. Requirements
|
||||
To run these examples with BigDL-LLM on Intel GPUs, we have some recommended requirements for your machine, please refer to [here](../README.md#recommended-requirements) for more information.
|
||||
|
||||
## Example: Predict Tokens using `generate()` API
|
||||
In the example [generate.py](./generate.py), we show a basic use case for a Baichuan model to predict the next N tokens using `generate()` API, with BigDL-LLM INT4 optimizations on Intel GPUs.
|
||||
### 1. Install
|
||||
We suggest using conda to manage environment:
|
||||
```bash
|
||||
conda create -n llm python=3.9
|
||||
conda activate llm
|
||||
# below command will install intel_extension_for_pytorch==2.0.110+xpu as default
|
||||
# you can install specific ipex/torch version for your need
|
||||
pip install --pre --upgrade bigdl-llm[xpu] -f https://developer.intel.com/ipex-whl-stable-xpu
|
||||
pip install transformers_stream_generator # additional package required for Baichuan-7B-Chat to conduct generation
|
||||
```
|
||||
|
||||
### 2. Configures OneAPI environment variables
|
||||
```bash
|
||||
source /opt/intel/oneapi/setvars.sh
|
||||
```
|
||||
|
||||
### 3. Run
|
||||
|
||||
For optimal performance on Arc, it is recommended to set several environment variables.
|
||||
|
||||
```bash
|
||||
export USE_XETLA=OFF
|
||||
export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
|
||||
```
|
||||
|
||||
```
|
||||
python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
|
||||
```
|
||||
|
||||
Arguments info:
|
||||
- `--repo-id-or-model-path REPO_ID_OR_MODEL_PATH`: argument defining the huggingface repo id for the Baichuan model (e.g `baichuan-inc/Baichuan2-7B-Chat`) to be downloaded, or the path to the huggingface checkpoint folder. It is default to be `'baichuan-inc/Baichuan2-7B-Chat'`.
|
||||
- `--prompt PROMPT`: argument defining the prompt to be infered (with integrated prompt format for chat). It is default to be `'AI是什么？'`.
|
||||
- `--n-predict N_PREDICT`: argument defining the max number of tokens to predict. It is default to be `32`.
|
||||
|
||||
#### Sample Output
|
||||
#### [baichuan-inc/Baichuan2-7B-Chat](https://huggingface.co/baichuan-inc/Baichuan2-7B-Chat)
|
||||
```log
|
||||
-------------------- Prompt --------------------
|
||||
<human>AI是什么？ <bot>
|
||||
-------------------- Output --------------------
|
||||
<human>AI是什么？ <bot>
|
||||
AI是人工智能（Artificial Intelligence）的缩写，它是指让计算机或机器模拟、扩展和辅助人类的智能。AI技术已经广泛应用于各个领域
|
||||
```
|
||||
|
||||
```log
|
||||
Inference time: xxxx s
|
||||
-------------------- Prompt --------------------
|
||||
<human>What is AI? <bot>
|
||||
-------------------- Output --------------------
|
||||
<human>What is AI? <bot>Artificial Intelligence (AI) refers to the development of computer systems that can perform tasks that would typically require human intelligence. These tasks include learning, reasoning, problem
|
||||
```
|
||||
|
|
@ -0,0 +1,78 @@
|
|||
#
|
||||
# Copyright 2016 The BigDL Authors.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
#
|
||||
|
||||
import torch
|
||||
import intel_extension_for_pytorch as ipex
|
||||
import time
|
||||
import argparse
|
||||
|
||||
from bigdl.llm.transformers import AutoModelForCausalLM
|
||||
from transformers import AutoTokenizer
|
||||
|
||||
# you could tune the prompt based on your own model,
|
||||
BAICHUAN_PROMPT_FORMAT = "<human>{prompt} <bot>"
|
||||
|
||||
if __name__ == '__main__':
|
||||
parser = argparse.ArgumentParser(description='Predict Tokens using `generate()` API for Baichuan model')
|
||||
parser.add_argument('--repo-id-or-model-path', type=str, default="baichuan-inc/Baichuan2-7B-Chat",
|
||||
help='The huggingface repo id for the Baichuan model to be downloaded'
|
||||
', or the path to the huggingface checkpoint folder')
|
||||
parser.add_argument('--prompt', type=str, default="AI是什么？",
|
||||
help='Prompt to infer')
|
||||
parser.add_argument('--n-predict', type=int, default=32,
|
||||
help='Max tokens to predict')
|
||||
|
||||
args = parser.parse_args()
|
||||
model_path = args.repo_id_or_model_path
|
||||
|
||||
# Load model in 4 bit,
|
||||
# which convert the relevant layers in the model into INT4 format
|
||||
# if your selected model is capable of utilizing previous key/value attentions
|
||||
# to enhance decoding speed, but has `"use_cache": false` in its model config,
|
||||
# it is important to set `use_cache=True` explicitly in the `generate` function
|
||||
# to obtain optimal performance with BigDL-LLM INT4 optimizations
|
||||
model = AutoModelForCausalLM.from_pretrained(model_path,
|
||||
load_in_4bit=True,
|
||||
optimize_model=False,
|
||||
trust_remote_code=True,
|
||||
use_cache=True)
|
||||
model = model.to('xpu')
|
||||
|
||||
# Load tokenizer
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_path,
|
||||
trust_remote_code=True)
|
||||
|
||||
# Generate predicted tokens
|
||||
with torch.inference_mode():
|
||||
prompt = BAICHUAN_PROMPT_FORMAT.format(prompt=args.prompt)
|
||||
input_ids = tokenizer.encode(prompt, return_tensors="pt").to('xpu')
|
||||
# ipex model needs a warmup, then inference time can be accurate
|
||||
output = model.generate(input_ids,
|
||||
max_new_tokens=args.n_predict)
|
||||
|
||||
# start inference
|
||||
st = time.time()
|
||||
output = model.generate(input_ids,
|
||||
max_new_tokens=args.n_predict)
|
||||
torch.xpu.synchronize()
|
||||
end = time.time()
|
||||
output = output.cpu()
|
||||
output_str = tokenizer.decode(output[0], skip_special_tokens=True)
|
||||
print(f'Inference time: {end-st} s')
|
||||
print('-'*20, 'Prompt', '-'*20)
|
||||
print(prompt)
|
||||
print('-'*20, 'Output', '-'*20)
|
||||
print(output_str)
|
||||
Loading…
Reference in a new issue