AI Lab Docker Environment

This repository provides a GPU-enabled Docker environment containing multiple isolated Python environments for different AI workflows.

The container is based on CUDA 12.1 + Ubuntu 22.04 and includes environments for:

• TensorFlow research
• PyTorch deep learning
• Data science workflows
• Resume OCR + ATS scoring with Transformers

---

Environments Overview

The container contains four virtual environments located in:

/opt/envs/

Environment	Path	Purpose
TensorFlow	/opt/envs/tf	TensorFlow experiments and research
PyTorch	/opt/envs/torch	Deep learning, computer vision, YOLO
Data Science	/opt/envs/ds	Classical ML and analytics
ATS Resume AI	/opt/envs/ats	OCR, NLP, resume parsing, ATS scoring

---

Build the Docker Image

Run the following command from the project directory:

docker build -t ai-lab:latest .

---

Run the Container

To start the container with GPU support:

docker run -it --gpus all \
-v $(pwd):/workspace \
ai-lab:latest

This mounts your current project folder to /workspace inside the container.

---

Activating Python Environments

Each environment can be activated manually using source.

---

TensorFlow Environment

Activate:

source /opt/envs/tf/bin/activate

Verify:

python -c "import tensorflow as tf; print(tf.__version__)"

Deactivate:

deactivate

---

PyTorch Environment

Activate:

source /opt/envs/torch/bin/activate

Verify GPU:

python -c "import torch; print(torch.cuda.is_available())"

Deactivate:

deactivate

---

Data Science Environment

Activate:

source /opt/envs/ds/bin/activate

Test:

python -c "import pandas, sklearn, xgboost"

Deactivate:

deactivate

---

ATS Resume AI Environment

This environment is designed for:

• Resume OCR
• EasyOCR
• Transformers
• Sentence embeddings
• ATS scoring

Activate:

source /opt/envs/ats/bin/activate

Test OCR:

python -c "import easyocr; print('EasyOCR ready')"

Deactivate:

deactivate

---

Running Python Without Activating

You can also directly call Python from an environment:

/opt/envs/torch/bin/python script.py

Example:

/opt/envs/ats/bin/python resume_parser.py

---

Available Jupyter Kernels

If using JupyterLab inside the container, the following kernels are available:

• Python (TensorFlow)
• Python (PyTorch)
• Python (Data Science)
• Python (ATS Resume AI)

Start JupyterLab:

/opt/envs/ds/bin/jupyter lab --ip=0.0.0.0 --port=8888 --allow-root

---

Workspace Directory

The container uses:

/workspace

as the main working directory.

Any files in your host project directory will appear here inside the container.

---

GPU Support

The container supports CUDA-enabled GPUs.

Verify inside the container:

nvidia-smi

or:

python -c "import torch; print(torch.cuda.is_available())"

---

Included Technologies

The container includes support for:

• TensorFlow
• PyTorch
• HuggingFace Transformers
• Sentence Transformers
• EasyOCR
• Tesseract OCR
• OpenCV
• YOLOv8
• Streamlit / Gradio / FastAPI
• Pandas / Scikit-learn / XGBoost
• JupyterLab

---

🌐 Running Streamlit Apps

The container supports running Streamlit applications for building interactive ML demos.

---

▶️ Run a Streamlit App

Activate the desired environment (example: Data Science):

source /opt/envs/ds/bin/activate

Run the app:

streamlit run app.py --server.address 0.0.0.0 --server.port 8501

---

🌍 Access the App

• Local:

  http://localhost:8501
  

• Remote (Coder / VM / Cloud):

  • Forward port 8501
  • Open the forwarded URL

---

⚠️ Notes

• 0.0.0.0 is required for Docker/remote environments
• Default port is 8501
• Change port if needed:

streamlit run app.py --server.address 0.0.0.0 --server.port 8080

---

📜 License

MIT License