ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

LLM: adjust dolly v2 GPU example README (#9318)

Browse source
This commit is contained in:
Jin Qiao 2023-11-01 09:50:22 +08:00 committed by GitHub
parent c44c6dc43a
commit 96f8158fe2
1 changed files with 18 additions and 27 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

45
python/llm/example/GPU/HF-Transformers-AutoModels/Model/dolly-v2/README.md
View file

@ -11,11 +11,25 @@ We suggest using conda to manage environment:
```bash ```bash
conda create -n llm python=3.9 conda create -n llm python=3.9
conda activate llm conda activate llm
# below command will install intel_extension_for_pytorch==2.0.110+xpu as default
pip install bigdl-llm[all] # install bigdl-llm with 'all' option # 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
```
### 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
``` ```
### 2. Run
``` ```
python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
``` ```
@ -25,30 +39,7 @@ Arguments info:
- `--prompt PROMPT`: argument defining the prompt to be infered (with integrated prompt format for chat). It is default to be `'What is AI?'`. - `--prompt PROMPT`: argument defining the prompt to be infered (with integrated prompt format for chat). It is default to be `'What is AI?'`.
- `--n-predict N_PREDICT`: argument defining the max number of tokens to predict. It is default to be `32`. - `--n-predict N_PREDICT`: argument defining the max number of tokens to predict. It is default to be `32`.
> **Note**: When loading the model in 4-bit, BigDL-LLM converts linear layers in the model into INT4 format. In theory, a *X*B model saved in 16-bit will requires approximately 2*X* GB of memory for loading, and ~0.5*X* GB memory for further inference. #### Sample Output
>
> Please select the appropriate size of the Dolly v2 model based on the capabilities of your machine.
#### 2.1 Client
On client Windows machine, it is recommended to run directly with full utilization of all cores:
```powershell
python ./generate.py
```
#### 2.2 Server
For optimal performance on server, it is recommended to set several environment variables (refer to [here](../README.md#best-known-configuration-on-linux) for more information), and run the example with all the physical cores of a single socket.
E.g. on Linux,
```bash
# set BigDL-Nano env variables
source bigdl-nano-init
# e.g. for a server with 48 cores per socket
export OMP_NUM_THREADS=48
numactl -C 0-47 -m 0 python ./generate.py
```
#### 2.3 Sample Output
#### [databricks/dolly-v2-12b](https://huggingface.co/databricks/dolly-v2-12b) #### [databricks/dolly-v2-12b](https://huggingface.co/databricks/dolly-v2-12b)
```log ```log
Inference time: xxxx s Inference time: xxxx s