Skip to content

GitLab

"doc/vscode:/vscode.git/clone" did not exist on "1b7672937b9ba679b1ad077c1760203aefb77226"
Commit 84987715 authored by chenych's avatar chenych
Browse files

update to v0.9.2

parent 317a82e2
Hide whitespace changes
Inline Side-by-side
Showing with 69 additions and 378 deletions
README_en.md
View file @ 84987715
......@@ -5,7 +5,7 @@
[![GitHub contributors](https://img.shields.io/github/contributors/hiyouga/LLaMA-Factory?color=orange)](https://github.com/hiyouga/LLaMA-Factory/graphs/contributors)
[![GitHub workflow](https://github.com/hiyouga/LLaMA-Factory/actions/workflows/tests.yml/badge.svg)](https://github.com/hiyouga/LLaMA-Factory/actions/workflows/tests.yml)
[![PyPI](https://img.shields.io/pypi/v/llamafactory)](https://pypi.org/project/llamafactory/)
[![Citation](https://img.shields.io/badge/citation-319-green)](https://scholar.google.com/scholar?cites=12620864006390196564)
[![Citation](https://img.shields.io/badge/citation-349-green)](https://scholar.google.com/scholar?cites=12620864006390196564)
[![GitHub pull request](https://img.shields.io/badge/PRs-welcome-blue)](https://github.com/hiyouga/LLaMA-Factory/pulls)
[![Twitter](https://img.shields.io/twitter/follow/llamafactory_ai)](https://twitter.com/llamafactory_ai)
......@@ -37,10 +37,10 @@ https://github.com/user-attachments/assets/7c96b465-9df7-45f4-8053-bf03e58386d3
Choose your path:
- **Documentation (WIP)**: https://llamafactory.readthedocs.io/zh-cn/latest/
- **Colab**: https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing
- **Documentation**: https://llamafactory.readthedocs.io/en/latest/
- **Colab (free)**: https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing
- **Local machine**: Please refer to [usage](#getting-started)
- **PAI-DSW**: [Llama3 Example](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory) | [Qwen2-VL Example](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_qwen2vl) | [DeepSeek-R1-Distill Example](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_deepseek_r1_distill_7b)
- **PAI-DSW (free trial)**: [Llama3 Example](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory) | [Qwen2-VL Example](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_qwen2vl) | [DeepSeek-R1-Distill Example](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_deepseek_r1_distill_7b)
- **Amazon SageMaker**: [Blog](https://aws.amazon.com/cn/blogs/china/a-one-stop-code-free-model-fine-tuning-deployment-platform-based-on-sagemaker-and-llama-factory/)
> [!NOTE]
......@@ -403,7 +403,7 @@ huggingface-cli login
| Optional | Minimum | Recommend |
| ------------ | ------- | --------- |
| CUDA | 11.6 | 12.2 |
| deepspeed | 0.10.0 | 0.16.2 |
| deepspeed | 0.10.0 | 0.16.4 |
| bitsandbytes | 0.39.0 | 0.43.1 |
| vllm | 0.4.3 | 0.7.3 |
| flash-attn | 2.3.0 | 2.7.2 |
......@@ -412,15 +412,14 @@ huggingface-cli login
\* *estimated*
| Method | Bits | 7B | 13B | 30B | 70B | 110B | 8x7B | 8x22B |
| ------------------------ | ---- | ----- | ----- | ----- | ------ | ------ | ----- | ------ |
| Full | 32 | 120GB | 240GB | 600GB | 1200GB | 2000GB | 900GB | 2400GB |
| Full | 16 | 60GB | 120GB | 300GB | 600GB | 900GB | 400GB | 1200GB |
| Freeze | 16 | 20GB | 40GB | 80GB | 200GB | 360GB | 160GB | 400GB |
| LoRA/GaLore/APOLLO/BAdam | 16 | 16GB | 32GB | 64GB | 160GB | 240GB | 120GB | 320GB |
| QLoRA | 8 | 10GB | 20GB | 40GB | 80GB | 140GB | 60GB | 160GB |
| QLoRA | 4 | 6GB | 12GB | 24GB | 48GB | 72GB | 30GB | 96GB |
| QLoRA | 2 | 4GB | 8GB | 16GB | 24GB | 48GB | 18GB | 48GB |
| Method | Bits | 7B | 14B | 30B | 70B | `x`B |
| ------------------------------- | ---- | ----- | ----- | ----- | ------ | ------- |
| Full (`bf16` or `fp16`) | 32 | 120GB | 240GB | 600GB | 1200GB | `18x`GB |
| Full (`pure_bf16`) | 16 | 60GB | 120GB | 300GB | 600GB | `8x`GB |
| Freeze/LoRA/GaLore/APOLLO/BAdam | 16 | 16GB | 32GB | 64GB | 160GB | `2x`GB |
| QLoRA | 8 | 10GB | 20GB | 40GB | 80GB | `x`GB |
| QLoRA | 4 | 6GB | 12GB | 24GB | 48GB | `x/2`GB |
| QLoRA | 2 | 4GB | 8GB | 16GB | 24GB | `x/4`GB |
## Getting Started
......@@ -490,12 +489,12 @@ bash Ascend-cann-kernels-910b_8.0.0.alpha002_linux-"$(uname -i)".run --install
source /usr/local/Ascend/ascend-toolkit/set_env.sh
```
| Requirement | Minimum | Recommend |
| ------------ | ------- | ----------- |
| CANN | 8.0.RC1 | 8.0.0.alpha002 |
| torch | 2.1.0 | 2.4.0 |
| torch-npu | 2.1.0 | 2.4.0.post2 |
| deepspeed | 0.13.2 | 0.16.2 |
| Requirement | Minimum | Recommend |
| ------------ | ------- | -------------- |
| CANN | 8.0.RC1 | 8.0.0.alpha002 |
| torch | 2.1.0 | 2.4.0 |
| torch-npu | 2.1.0 | 2.4.0.post2 |
| deepspeed | 0.13.2 | 0.13.2 |
Remember to use `ASCEND_RT_VISIBLE_DEVICES` instead of `CUDA_VISIBLE_DEVICES` to specify the device to use.
......@@ -560,6 +559,8 @@ See [examples/README.md](examples/README.md) for advanced usage (including distr
> [!TIP]
> Use `llamafactory-cli help` to show help information.
>
> Read [FAQs](https://github.com/hiyouga/LLaMA-Factory/issues/4614) first if you encounter any problems.
### Fine-Tuning with LLaMA Board GUI (powered by [Gradio](https://github.com/gradio-app/gradio))
......
README_zh.md
View file @ 84987715
......@@ -5,7 +5,7 @@
[![GitHub contributors](https://img.shields.io/github/contributors/hiyouga/LLaMA-Factory?color=orange)](https://github.com/hiyouga/LLaMA-Factory/graphs/contributors)
[![GitHub workflow](https://github.com/hiyouga/LLaMA-Factory/actions/workflows/tests.yml/badge.svg)](https://github.com/hiyouga/LLaMA-Factory/actions/workflows/tests.yml)
[![PyPI](https://img.shields.io/pypi/v/llamafactory)](https://pypi.org/project/llamafactory/)
[![Citation](https://img.shields.io/badge/citation-319-green)](https://scholar.google.com/scholar?cites=12620864006390196564)
[![Citation](https://img.shields.io/badge/citation-349-green)](https://scholar.google.com/scholar?cites=12620864006390196564)
[![GitHub pull request](https://img.shields.io/badge/PRs-welcome-blue)](https://github.com/hiyouga/LLaMA-Factory/pulls)
[![Twitter](https://img.shields.io/twitter/follow/llamafactory_ai)](https://twitter.com/llamafactory_ai)
......@@ -40,9 +40,9 @@ https://github.com/user-attachments/assets/e6ce34b0-52d5-4f3e-a830-592106c4c272
- **入门教程**：https://zhuanlan.zhihu.com/p/695287607
- **框架文档**：https://llamafactory.readthedocs.io/zh-cn/latest/
- **Colab**：https://colab.research.google.com/drive/1d5KQtbemerlSDSxZIfAaWXhKr30QypiK?usp=sharing
- **Colab（免费）**：https://colab.research.google.com/drive/1d5KQtbemerlSDSxZIfAaWXhKr30QypiK?usp=sharing
- **本地机器**：请见[如何使用](#如何使用)
- **PAI-DSW**：[Llama3 案例](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory) | [Qwen2-VL 案例](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_qwen2vl) | [DeepSeek-R1-Distill 案例](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_deepseek_r1_distill_7b)
- **PAI-DSW（免费试用）**：[Llama3 案例](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory) | [Qwen2-VL 案例](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_qwen2vl) | [DeepSeek-R1-Distill 案例](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory_deepseek_r1_distill_7b)
- **Amazon SageMaker**：[博客](https://aws.amazon.com/cn/blogs/china/a-one-stop-code-free-model-fine-tuning-deployment-platform-based-on-sagemaker-and-llama-factory/)
> [!NOTE]
......@@ -405,7 +405,7 @@ huggingface-cli login
| 可选项 | 至少 | 推荐 |
| ------------ | ------- | --------- |
| CUDA | 11.6 | 12.2 |
| deepspeed | 0.10.0 | 0.16.2 |
| deepspeed | 0.10.0 | 0.16.4 |
| bitsandbytes | 0.39.0 | 0.43.1 |
| vllm | 0.4.3 | 0.7.3 |
| flash-attn | 2.3.0 | 2.7.2 |
......@@ -414,15 +414,14 @@ huggingface-cli login
\* *估算值*
| 方法 | 精度 | 7B | 13B | 30B | 70B | 110B | 8x7B | 8x22B |
| ------------------------ | ---- | ----- | ----- | ----- | ------ | ------ | ----- | ------ |
| Full | 32 | 120GB | 240GB | 600GB | 1200GB | 2000GB | 900GB | 2400GB |
| Full | 16 | 60GB | 120GB | 300GB | 600GB | 900GB | 400GB | 1200GB |
| Freeze | 16 | 20GB | 40GB | 80GB | 200GB | 360GB | 160GB | 400GB |
| LoRA/GaLore/APOLLO/BAdam | 16 | 16GB | 32GB | 64GB | 160GB | 240GB | 120GB | 320GB |
| QLoRA | 8 | 10GB | 20GB | 40GB | 80GB | 140GB | 60GB | 160GB |
| QLoRA | 4 | 6GB | 12GB | 24GB | 48GB | 72GB | 30GB | 96GB |
| QLoRA | 2 | 4GB | 8GB | 16GB | 24GB | 48GB | 18GB | 48GB |
| 方法 | 精度 | 7B | 14B | 30B | 70B | `x`B |
| ------------------------------- | ---- | ----- | ----- | ----- | ------ | ------- |
| Full (`bf16` or `fp16`) | 32 | 120GB | 240GB | 600GB | 1200GB | `18x`GB |
| Full (`pure_bf16`) | 16 | 60GB | 120GB | 300GB | 600GB | `8x`GB |
| Freeze/LoRA/GaLore/APOLLO/BAdam | 16 | 16GB | 32GB | 64GB | 160GB | `2x`GB |
| QLoRA | 8 | 10GB | 20GB | 40GB | 80GB | `x`GB |
| QLoRA | 4 | 6GB | 12GB | 24GB | 48GB | `x/2`GB |
| QLoRA | 2 | 4GB | 8GB | 16GB | 24GB | `x/4`GB |
## 如何使用
......@@ -493,12 +492,12 @@ bash Ascend-cann-kernels-910b_8.0.RC1.alpha001_linux.run --install
source /usr/local/Ascend/ascend-toolkit/set_env.sh
```
| 依赖项 | 至少 | 推荐 |
| ------------ | ------- | ----------- |
| CANN | 8.0.RC1 | 8.0.RC1 |
| torch | 2.1.0 | 2.1.0 |
| torch-npu | 2.1.0 | 2.1.0.post3 |
| deepspeed | 0.13.2 | 0.13.2 |
| 依赖项 | 至少 | 推荐 |
| ------------ | ------- | -------------- |
| CANN | 8.0.RC1 | 8.0.0.alpha002 |
| torch | 2.1.0 | 2.4.0 |
| torch-npu | 2.1.0 | 2.4.0.post2 |
| deepspeed | 0.13.2 | 0.13.2 |
请使用 `ASCEND_RT_VISIBLE_DEVICES` 而非 `CUDA_VISIBLE_DEVICES` 来指定运算设备。
......@@ -563,6 +562,8 @@ llamafactory-cli export examples/merge_lora/llama3_lora_sft.yaml
> [!TIP]
> 使用 `llamafactory-cli help` 显示帮助信息。
>
> 遇到报错请先看[常见问题](https://github.com/hiyouga/LLaMA-Factory/issues/4614)。
### LLaMA Board 可视化微调（由 [Gradio](https://github.com/gradio-app/gradio) 驱动）
......
assets/wechat.jpg
View replaced file @ 317a82e2 View file @ 84987715
assets/wechat.jpg

167 KB | W: | H:

assets/wechat.jpg

164 KB | W: | H:

assets/wechat.jpg
assets/wechat.jpg
assets/wechat.jpg
assets/wechat.jpg
  • 2-up
  • Swipe
  • Onion skin
assets/wechat_npu.jpg
View replaced file @ 317a82e2 View file @ 84987715
assets/wechat_npu.jpg

167 KB | W: | H:

assets/wechat_npu.jpg

167 KB | W: | H:

assets/wechat_npu.jpg
assets/wechat_npu.jpg
assets/wechat_npu.jpg
assets/wechat_npu.jpg
  • 2-up
  • Swipe
  • Onion skin
data/mllm_demo.json
View file @ 84987715
......@@ -10,7 +10,7 @@
"role": "assistant"
},
{
"content": "What are they doing?",
"content": "What are they doing?<image>",
"role": "user"
},
{
......@@ -19,6 +19,7 @@
}
],
"images": [
"mllm_demo_data/1.jpg",
"mllm_demo_data/1.jpg"
]
},
......@@ -79,7 +80,7 @@
"role": "assistant"
},
{
"content": "他们在做什么？",
"content": "他们在做什么？<image>",
"role": "user"
},
{
......@@ -88,6 +89,7 @@
}
],
"images": [
"mllm_demo_data/1.jpg",
"mllm_demo_data/1.jpg"
]
},
......
docker/docker-cuda/Dockerfile deleted 100644 → 0
View file @ 317a82e2
# Default use the NVIDIA official image with PyTorch 2.3.0
# https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/index.html
ARG BASE_IMAGE=nvcr.io/nvidia/pytorch:24.02-py3
FROM ${BASE_IMAGE}
# Define environments
ENV MAX_JOBS=4
ENV FLASH_ATTENTION_FORCE_BUILD=TRUE
ENV VLLM_WORKER_MULTIPROC_METHOD=spawn
# Define installation arguments
ARG INSTALL_BNB=false
ARG INSTALL_VLLM=false
ARG INSTALL_DEEPSPEED=false
ARG INSTALL_FLASHATTN=false
ARG INSTALL_LIGER_KERNEL=false
ARG INSTALL_HQQ=false
ARG INSTALL_EETQ=false
ARG PIP_INDEX=https://pypi.org/simple
ARG HTTP_PROXY=
# Set the working directory
WORKDIR /app
# Set http proxy
RUN if [ -n "$HTTP_PROXY" ]; then \
echo "Configuring proxy..."; \
export http_proxy=$HTTP_PROXY; \
export https_proxy=$HTTP_PROXY; \
fi
# Install the requirements
COPY requirements.txt /app
RUN pip config set global.index-url "$PIP_INDEX" && \
pip config set global.extra-index-url "$PIP_INDEX" && \
python -m pip install --upgrade pip && \
if [ -n "$HTTP_PROXY" ]; then \
python -m pip install --proxy=$HTTP_PROXY -r requirements.txt; \
else \
python -m pip install -r requirements.txt; \
fi
# Copy the rest of the application into the image
COPY . /app
# Install the LLaMA Factory
RUN EXTRA_PACKAGES="metrics"; \
if [ "$INSTALL_BNB" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},bitsandbytes"; \
fi; \
if [ "$INSTALL_VLLM" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},vllm"; \
fi; \
if [ "$INSTALL_DEEPSPEED" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},deepspeed"; \
fi; \
if [ "$INSTALL_LIGER_KERNEL" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},liger-kernel"; \
fi; \
if [ "$INSTALL_HQQ" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},hqq"; \
fi; \
if [ "$INSTALL_EETQ" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},eetq"; \
fi; \
if [ -n "$HTTP_PROXY" ]; then \
pip install --proxy=$HTTP_PROXY -e ".[$EXTRA_PACKAGES]"; \
else \
pip install -e ".[$EXTRA_PACKAGES]"; \
fi
# Rebuild flash attention
RUN pip uninstall -y transformer-engine flash-attn && \
if [ "$INSTALL_FLASHATTN" == "true" ]; then \
pip uninstall -y ninja && \
if [ -n "$HTTP_PROXY" ]; then \
pip install --proxy=$HTTP_PROXY ninja && \
pip install --proxy=$HTTP_PROXY --no-cache-dir flash-attn --no-build-isolation; \
else \
pip install ninja && \
pip install --no-cache-dir flash-attn --no-build-isolation; \
fi; \
fi
# Unset http proxy
RUN if [ -n "$HTTP_PROXY" ]; then \
unset http_proxy; \
unset https_proxy; \
fi
# Set up volumes
VOLUME [ "/root/.cache/huggingface", "/root/.cache/modelscope", "/app/data", "/app/output" ]
# Expose port 7860 for the LLaMA Board
ENV GRADIO_SERVER_PORT 7860
EXPOSE 7860
# Expose port 8000 for the API service
ENV API_PORT 8000
EXPOSE 8000
docker/docker-cuda/docker-compose.yml deleted 100644 → 0
View file @ 317a82e2
services:
llamafactory:
build:
dockerfile: ./docker/docker-cuda/Dockerfile
context: ../..
args:
INSTALL_BNB: false
INSTALL_VLLM: false
INSTALL_DEEPSPEED: false
INSTALL_FLASHATTN: false
INSTALL_LIGER_KERNEL: false
INSTALL_HQQ: false
INSTALL_EETQ: false
PIP_INDEX: https://pypi.org/simple
container_name: llamafactory
volumes:
- ../../hf_cache:/root/.cache/huggingface
- ../../ms_cache:/root/.cache/modelscope
- ../../om_cache:/root/.cache/openmind
- ../../data:/app/data
- ../../output:/app/output
ports:
- "7860:7860"
- "8000:8000"
ipc: host
tty: true
shm_size: '16gb'
stdin_open: true
command: bash
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: "all"
capabilities: [gpu]
restart: unless-stopped
docker/docker-npu/Dockerfile deleted 100644 → 0
View file @ 317a82e2
# Use the Ubuntu 22.04 image with CANN 8.0.rc1
# More versions can be found at https://hub.docker.com/r/ascendai/cann/tags
# FROM ascendai/cann:8.0.rc1-910-ubuntu22.04-py3.8
FROM ascendai/cann:8.0.0-910b-ubuntu22.04-py3.10
# FROM ascendai/cann:8.0.rc1-910-openeuler22.03-py3.8
# FROM ascendai/cann:8.0.rc1-910b-openeuler22.03-py3.8
# Define environments
ENV DEBIAN_FRONTEND=noninteractive
# Define installation arguments
ARG INSTALL_DEEPSPEED=false
ARG PIP_INDEX=https://pypi.org/simple
ARG TORCH_INDEX=https://download.pytorch.org/whl/cpu
ARG HTTP_PROXY=
# Set the working directory
WORKDIR /app
# Set http proxy
RUN if [ -n "$HTTP_PROXY" ]; then \
echo "Configuring proxy..."; \
export http_proxy=$HTTP_PROXY; \
export https_proxy=$HTTP_PROXY; \
fi
# Install the requirements
COPY requirements.txt /app
RUN pip config set global.index-url "$PIP_INDEX" && \
pip config set global.extra-index-url "$TORCH_INDEX" && \
python -m pip install --upgrade pip && \
if [ -n "$HTTP_PROXY" ]; then \
python -m pip install --proxy=$HTTP_PROXY -r requirements.txt; \
else \
python -m pip install -r requirements.txt; \
fi
# Copy the rest of the application into the image
COPY . /app
# Install the LLaMA Factory
RUN EXTRA_PACKAGES="torch-npu,metrics"; \
if [ "$INSTALL_DEEPSPEED" == "true" ]; then \
EXTRA_PACKAGES="${EXTRA_PACKAGES},deepspeed"; \
fi; \
if [ -n "$HTTP_PROXY" ]; then \
pip install --proxy=$HTTP_PROXY -e ".[$EXTRA_PACKAGES]"; \
else \
pip install -e ".[$EXTRA_PACKAGES]"; \
fi
# Unset http proxy
RUN if [ -n "$HTTP_PROXY" ]; then \
unset http_proxy; \
unset https_proxy; \
fi
# Set up volumes
VOLUME [ "/root/.cache/huggingface", "/root/.cache/modelscope", "/app/data", "/app/output" ]
# Expose port 7860 for the LLaMA Board
ENV GRADIO_SERVER_PORT 7860
EXPOSE 7860
# Expose port 8000 for the API service
ENV API_PORT 8000
EXPOSE 8000
docker/docker-npu/docker-compose.yml deleted 100644 → 0
View file @ 317a82e2
services:
llamafactory:
build:
dockerfile: ./docker/docker-npu/Dockerfile
context: ../..
args:
INSTALL_DEEPSPEED: "false"
PIP_INDEX: https://pypi.org/simple
container_name: llamafactory
volumes:
- ../../hf_cache:/root/.cache/huggingface
- ../../ms_cache:/root/.cache/modelscope
- ../../om_cache:/root/.cache/openmind
- ../../data:/app/data
- ../../output:/app/output
- /usr/local/dcmi:/usr/local/dcmi
- /usr/local/bin/npu-smi:/usr/local/bin/npu-smi
- /usr/local/Ascend/driver:/usr/local/Ascend/driver
- /etc/ascend_install.info:/etc/ascend_install.info
ports:
- "7860:7860"
- "8000:8000"
ipc: host
tty: true
shm_size: '16gb'
stdin_open: true
command: bash
devices:
- /dev/davinci0
- /dev/davinci_manager
- /dev/devmm_svm
- /dev/hisi_hdc
restart: unless-stopped
docker/docker-rocm/docker-compose.yml
View file @ 84987715
......@@ -4,12 +4,12 @@ services:
dockerfile: ./docker/docker-rocm/Dockerfile
context: ../..
args:
INSTALL_BNB: false
INSTALL_VLLM: false
INSTALL_DEEPSPEED: false
INSTALL_FLASHATTN: false
INSTALL_LIGER_KERNEL: false
INSTALL_HQQ: false
INSTALL_BNB: "false"
INSTALL_VLLM: "false"
INSTALL_DEEPSPEED: "false"
INSTALL_FLASHATTN: "false"
INSTALL_LIGER_KERNEL: "false"
INSTALL_HQQ: "false"
PIP_INDEX: https://pypi.org/simple
container_name: llamafactory
volumes:
......@@ -24,7 +24,7 @@ services:
- "8000:8000"
ipc: host
tty: true
shm_size: '16gb'
shm_size: "16gb"
stdin_open: true
command: bash
devices:
......
examples/accelerate/fsdp_config.yaml
View file @ 84987715
......@@ -14,7 +14,7 @@ fsdp_config:
fsdp_use_orig_params: true
machine_rank: 0
main_training_function: main
mixed_precision: fp16 # or bf16
mixed_precision: bf16 # or fp16
num_machines: 1 # the number of nodes
num_processes: 2 # the number of GPUs in all nodes
rdzv_backend: static
......
examples/deepspeed/ds_z0_config.json
View file @ 84987715
......@@ -19,7 +19,7 @@
"stage": 0,
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": true,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients": true,
......
examples/deepspeed/ds_z2_config.json
View file @ 84987715
......@@ -19,7 +19,7 @@
"stage": 2,
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": true,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients": true,
......
examples/deepspeed/ds_z2_offload_config.json
View file @ 84987715
......@@ -23,7 +23,7 @@
},
"allgather_partitions": true,
"allgather_bucket_size": 5e8,
"overlap_comm": true,
"overlap_comm": false,
"reduce_scatter": true,
"reduce_bucket_size": 5e8,
"contiguous_gradients": true,
......
examples/deepspeed/ds_z3_config.json
View file @ 84987715
......@@ -17,7 +17,7 @@
},
"zero_optimization": {
"stage": 3,
"overlap_comm": true,
"overlap_comm": false,
"contiguous_gradients": true,
"sub_group_size": 1e9,
"reduce_bucket_size": "auto",
......
examples/deepspeed/ds_z3_offload_config.json
View file @ 84987715
......@@ -25,7 +25,7 @@
"device": "cpu",
"pin_memory": true
},
"overlap_comm": true,
"overlap_comm": false,
"contiguous_gradients": true,
"sub_group_size": 1e9,
"reduce_bucket_size": "auto",
......
examples/train_full/llama3_full_sft_ds3.yaml → examples/train_full/llama3_full_sft.yaml
View file @ 84987715
### model
model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct
trust_remote_code: true
### method
stage: sft
do_train: true
finetuning_type: full
deepspeed: examples/deepspeed/ds_z3_config.json
deepspeed: examples/deepspeed/ds_z3_config.json # choices: [ds_z0_config.json, ds_z2_config.json, ds_z3_config.json]
### dataset
dataset: identity,alpaca_en_demo
......@@ -14,6 +15,7 @@ cutoff_len: 2048
max_samples: 1000
overwrite_cache: true
preprocessing_num_workers: 16
dataloader_num_workers: 4
### output
output_dir: saves/llama3-8b/full/sft
......@@ -21,6 +23,7 @@ logging_steps: 10
save_steps: 500
plot_loss: true
overwrite_output_dir: true
save_only_model: false
### train
per_device_train_batch_size: 1
......@@ -31,9 +34,11 @@ lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: true
ddp_timeout: 180000000
resume_from_checkpoint: null
### eval
val_size: 0.1
per_device_eval_batch_size: 1
eval_strategy: steps
eval_steps: 500
# eval_dataset: alpaca_en_demo
# val_size: 0.1
# per_device_eval_batch_size: 1
# eval_strategy: steps
# eval_steps: 500
examples/train_full/qwen2vl_full_sft.yaml
View file @ 84987715
......@@ -10,7 +10,7 @@ do_train: true
finetuning_type: full
freeze_vision_tower: true # choices: [true, false]
freeze_multi_modal_projector: true # choices: [true, false]
train_mm_proj_only: false # choices: [true, false]
freeze_language_model: false # choices: [true, false]
deepspeed: examples/deepspeed/ds_z3_config.json # choices: [ds_z0_config.json, ds_z2_config.json, ds_z3_config.json]
### dataset
......
examples/train_lora/qwen2.5_lora_sft_ds3.yaml deleted 100644 → 0
View file @ 317a82e2
### model
model_name_or_path: /data/luopl/Qwen/Qwen2.5-72B
### method
stage: sft
do_train: true
finetuning_type: lora
lora_target: q_proj,v_proj
deepspeed: examples/deepspeed/ds_z3_config.json
### dataset
dataset: identity,alpaca_zh_demo,alpaca_en_demo
template: qwen
cutoff_len: 1024
max_samples: 1000
overwrite_cache: true
preprocessing_num_workers: 4
### output
output_dir: saves/qwen2.5_72b/lora/sft/
logging_steps: 10
save_steps: 500
plot_loss: true
overwrite_output_dir: true
### train
per_device_train_batch_size: 1
gradient_accumulation_steps: 1
learning_rate: 1.0e-5
num_train_epochs: 3.0
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: true
ddp_timeout: 180000000
### eval
val_size: 0.1
per_device_eval_batch_size: 1
eval_strategy: steps
eval_steps: 250
examples/train_lora/qwen2.5_lora_sft_offload_ds3.yaml deleted 100644 → 0
View file @ 317a82e2
### model
model_name_or_path: /data/luopl/Qwen/Qwen2.5-72B
### method
stage: sft
do_train: true
finetuning_type: lora
lora_target: q_proj,v_proj
deepspeed: examples/deepspeed/ds_z3_offload_config.json
### dataset
dataset: identity,alpaca_zh_demo,alpaca_en_demo
template: qwen
cutoff_len: 1024
max_samples: 1000
overwrite_cache: true
preprocessing_num_workers: 4
### output
output_dir: saves/qwen2.5_72b/lora/sft/
logging_steps: 10
save_steps: 500
plot_loss: true
overwrite_output_dir: true
### train
per_device_train_batch_size: 1
gradient_accumulation_steps: 1
learning_rate: 1.0e-5
num_train_epochs: 3.0
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: true
ddp_timeout: 180000000
### eval
val_size: 0.1
per_device_eval_batch_size: 1
eval_strategy: steps
eval_steps: 250
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment