📝 docs: 更新文档

docs/dataset.md

@@ -15,6 +15,8 @@ outline: deep
 DATASET_DIR=/path/to/dataset
 ```
 
+> 需要注意的是，数据集通常需要放置在项目外的路径，例如 `~/data` （推荐）。如果放置在项目内，导致编辑器对于项目的索引过大，会导致编辑器卡顿
+
 这里准备好了一个示例数据集，可以下载
 
 ```bash

docs/install.md

@@ -38,29 +38,52 @@ git submodule init
 git submodule update
 ```
 
-### 创建虚拟环境
+### 创建环境并安装依赖
 
 确保安装了 conda ，如果没有安装，请从 [Miniconda](https://docs.anaconda.com/free/miniconda/index.html) 下载，或者快速安装
-  
+
 ```shell
 # linux x64
 wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
 bash Miniconda3-latest-Linux-x86_64.sh
 ```
 
+
+#### 方法一：使用提供的脚本
+
+
+查看 `scripts/base.sh` ，根据需要修改配置，执行命令自动创建并且激活虚拟环境
+
+::: code-group
+
+```shell [使用 venv 创建虚拟环境]
+bash scripts/python-activate.venv.sh
+```
+
+```shell [使用 conda 创建虚拟环境]
+bash scripts/python-activate.conda.sh
+```
+
+:::
+
+- **可以重复执行该脚本获取激活环境的提示信息或者安装依赖**
+- 修改 `export PIP_QUIET=true` 为 `false` 可以查看安装过程
+- 该脚本会复制 `yolov5/requirements.txt` 到 `.cache/yolov5/requirements.txt`，可以自行修改 `.cache/yolov5/requirements.txt` 文件安装相关依赖，例如取消 `onnx` 的注释以支持 ONNX 格式的模型导出
+
+#### 方法二：手动安装
+
 创建虚拟环境
 ```shell
+export ENV_NAME=deep-object-detect-track
 # 在项目内安装环境(推荐)
-conda create -p .env/dodt python=3.10 -y
-conda activate ./.env/dodt
+conda create -p .env/$ENV_NAME python=3.10 -y
+conda activate ./.env/$ENV_NAME
 # 全局安装环境
-conda create -n dodt python=3.10 -y
-conda activate dodt
+conda create -n $ENV_NAME python=3.10 -y
+conda activate $ENV_NAME
 ```
 > Python 版本选择 3.10 是因为 Ubuntu 22.04 默认安装的 Python 版本是 3.10
 
-### 安装依赖
-
 1. 安装 PyTorch
 
 参考官网 [INSTALL PYTORCH](https://pytorch.org/get-started/locally/) 选择配置安装 PyTorch
@@ -75,7 +98,7 @@ pip install torch torchvision \
 
 ```shell
 pip install -r projects/yolov5/requirements.txt
-pip install -r requirements.txt
+pip install -r requirements/requirements.train.txt
 ```
 
 如果涉及部署流程，需要自行修改 `requirements.txt` 文件，将下列依赖取消注释掉，然后重新执行上述命令

scripts/base.sh

@@ -15,7 +15,7 @@ export ENV_PATH=$BASE_ENV_PATH/.env/$ENV_NAME
 
 # ================== Python Variables ==================
 CUSTOM_PYTHON_VERSION=3.12    #  Uncomment and set to the desired Python version
-export PIP_QUIET=false
+export PIP_QUIET=true
 
 # ================== Enabling OpenVINO ==================
 export OPENVINO_HOME=/opt/intel/openvino_2024

scripts/python-install-requirements.sh

@@ -31,27 +31,23 @@ if [ ! -f ".cache/yolov5/requirements.txt" ]; then
     cp projects/yolov5/requirements.txt .cache/yolov5/requirements.txt
 fi
 python3 -m pip install $PIP_QUIET_FLAG -r .cache/yolov5/requirements.txt
-python3 -m pip install $PIP_QUIET_FLAG -r requirements/requirements.txt
-
-python3 -m pip install $PIP_QUIET_FLAG onnx
-python3 -m pip install $PIP_QUIET_FLAG onnx-simplifier
+python3 -m pip install $PIP_QUIET_FLAG -r requirements/requirements.train.txt
 
 if [ ! -z "${CUDA_VERSION}" ]; then
     python3 -m pip install $PIP_QUIET_FLAG onnxruntime-gpu
 else
     python3 -m pip install $PIP_QUIET_FLAG onnxruntime
 fi
 
-# if [ -d "$INTEL_OPENVINO_DIR" ]; then
-#     print_info "OpenVINO found in $INTEL_OPENVINO_DIR"
-#     print_info "Installing OpenVINO Python requirements from $INTEL_OPENVINO_DIR/python/requirements.txt"
-#     python3 -m pip install $PIP_QUIET_FLAG -r $INTEL_OPENVINO_DIR/python/requirements.txt
-# else
-#     print_warning "OpenVINO not found"
-#     print_info "Installing OpenVINO Python requirements from pip"
-#     python3 -m pip install $PIP_QUIET_FLAG openvino-dev
-# fi
-
+if [ -d "$INTEL_OPENVINO_DIR" ]; then
+    print_info "OpenVINO found in $INTEL_OPENVINO_DIR"
+    print_info "Installing OpenVINO Python requirements from $INTEL_OPENVINO_DIR/python/requirements.txt"
+    python3 -m pip install $PIP_QUIET_FLAG -r $INTEL_OPENVINO_DIR/python/requirements.txt
+else
+    print_warning "OpenVINO not found, skipping OpenVINO Python requirements installation"
+    # print_info "Installing OpenVINO Python requirements from pip"
+    # python3 -m pip install $PIP_QUIET_FLAG openvino-dev
+fi
 
 # freeze the requirements 
 # python3 -m pip list --format=freeze > requirements.version.txt