I’m building an AWS Lambda with Python 3.10 & detectron2 in it and it’s not working at all.

# Use the Python 3.10 base image for AWS Lambda

```
FROM public.ecr.aws/lambda/python:3.10 AS build
```

# Set environment variables

```
ENV PYTHONDONTWRITEBYTECODE 1
ENV PYTHONUNBUFFERED 1
ENV LAMBDA_TASK_ROOT /var/task
```

# Set working directory

`WORKDIR ${LAMBDA_TASK_ROOT} `

# Install system dependencies for detectron2 and other required libraries

```
RUN yum update -y 
&& yum install -y 
git 
gcc 
gcc-c++ 
make 
cmake 
wget 
libpng-devel 
libjpeg-devel 
libtiff-devel 
zlib-devel 
mesa-libGL 
mesa-libGL-devel 
opencv 
tar 
&& yum clean all
```

# Install pip and upgrade it to the latest version

`RUN pip install --upgrade pip `

# Step 1: Install PyTorch and torchvision globally first so that detectron2 can find torch

`RUN pip install torch torchvision `

# Step 2: Install detectron2 from GitHub after torch is installed

`RUN pip install 'git+https://github.com/facebookresearch/detectron2.git' `

# Step 3: Install other required dependencies (cython, pycocotools, etc.)

`RUN pip install cython pycocotools `

# Step 4: Move all installed packages into ${LAMBDA_TASK_ROOT}

`RUN pip install --target "${LAMBDA_TASK_ROOT}" torch torchvision detectron2 cython pycocotools `

# Step 5: Install additional dependencies from requirements.txt

` COPY requirements.txt . RUN pip install --target "${LAMBDA_TASK_ROOT}" -r requirements.txt`

# Clean up the build to reduce image size

`RUN rm -rf /var/cache/yum && rm -rf /tmp/* `

# Expose necessary ports (optional)

`EXPOSE 8080 `

# Command to run the Lambda function

`CMD ["app.lambda_handler"] 

This is what my dockerfile looks like but currently im facing

"No module named torch"

An error in step 2, which is installing detectron2. Is there anyone who has already built this thing, kindly share your thoughts.