上传setup_and_upload.sh.js

#!/usr/bin/env bash
# setup_and_upload.sh
# 一键完成仓库克隆、编译、依赖安装、Git LFS 配置、模型权重上传和服务启动
# 使用前：替换 YOUR_USERNAME、YOUR_SECURE_TOKEN 和 /path/to/llama2-7b.gguf

set -e

# 1. 克隆仓库并初始化子模块
REPO_URL="https://github.com/your-username/my-chatbot-llama2-7b.git"
BRANCH="main"

echo "🚀 克隆仓库并初始化子模块"
git clone "$REPO_URL"
cd my-chatbot-llama2-7b
git checkout "$BRANCH"
git submodule update --init --recursive

# 2. 编译 llama.cpp 并安装 Python 依赖
echo "🔧 编译 llama.cpp 并安装依赖"
cd llama.cpp && make && cd ..
pip install -r requirements.txt

# 3. 安装并配置 Git LFS
echo "📦 安装并配置 Git LFS"
git lfs install
git lfs track "models/*.gguf"

# 4. 放置并上传模型权重
echo "💾 放置并上传模型权重"
mkdir -p models
cp /path/to/llama2-7b.gguf models/
git add .gitattributes models/llama2-7b.gguf
git commit -m "chore: add LLaMA2-7B model weights via LFS"
git push origin "$BRANCH"

# 5. 启动推理服务
echo "🚀 启动推理服务"
uvicorn app:app --host 0.0.0.0 --port 8000 --reload &

# 6. 测试模型推理
echo "✅ 部署完成！使用以下命令进行测试："
echo "curl -X POST http://localhost:8000/generate -H 'Content-Type: application/json' -d '{\"prompt\": \"你好，世界！\", \"token\": \"YOUR_SECURE_TOKEN\"}'"

setup_and_upload.sh.js

@@ -0,0 +1,42 @@
+#!/usr/bin/env bash
+# setup_and_upload.sh
+# 一键完成仓库克隆、编译、依赖安装、Git LFS 配置、模型权重上传和服务启动
+# 使用前：替换 YOUR_USERNAME、YOUR_SECURE_TOKEN 和 /path/to/llama2-7b.gguf
+
+set -e
+
+# 1. 克隆仓库并初始化子模块
+REPO_URL="https://github.com/your-username/my-chatbot-llama2-7b.git"
+BRANCH="main"
+
+echo "🚀 克隆仓库并初始化子模块"
+git clone "$REPO_URL"
+cd my-chatbot-llama2-7b
+git checkout "$BRANCH"
+git submodule update --init --recursive
+
+# 2. 编译 llama.cpp 并安装 Python 依赖
+echo "🔧 编译 llama.cpp 并安装依赖"
+cd llama.cpp && make && cd ..
+pip install -r requirements.txt
+
+# 3. 安装并配置 Git LFS
+echo "📦 安装并配置 Git LFS"
+git lfs install
+git lfs track "models/*.gguf"
+
+# 4. 放置并上传模型权重
+echo "💾 放置并上传模型权重"
+mkdir -p models
+cp /path/to/llama2-7b.gguf models/
+git add .gitattributes models/llama2-7b.gguf
+git commit -m "chore: add LLaMA2-7B model weights via LFS"
+git push origin "$BRANCH"
+
+# 5. 启动推理服务
+echo "🚀 启动推理服务"
+uvicorn app:app --host 0.0.0.0 --port 8000 --reload &
+
+# 6. 测试模型推理
+echo "✅ 部署完成！使用以下命令进行测试："
+echo "curl -X POST http://localhost:8000/generate -H 'Content-Type: application/json' -d '{\"prompt\": \"你好，世界！\", \"token\": \"YOUR_SECURE_TOKEN\"}'"