debiasing using k-steering

.gitignore

@@ -0,0 +1,4 @@
+
+# Data directory
+data/
+

INSTALL.md

@@ -0,0 +1,134 @@
+# Installation Guide - T2I Interpretability Toolkit
+
+## Quick Install
+
+### 1. Install Python Requirements
+
+```bash
+cd /home/ubuntu/T2I_Interp_toolkit
+pip install -r requirements.txt
+```
+
+### 2. Install the Toolkit (Development Mode)
+
+```bash
+# Install the main toolkit
+cd /home/ubuntu/T2I_Interp_toolkit
+pip install -e .
+
+# Install the dictionary learning submodule
+cd dictionary_learning
+pip install -e .
+```
+
+## Alternative: Install with Poetry
+
+If you prefer using Poetry for the dictionary learning module:
+
+```bash
+cd /home/ubuntu/T2I_Interp_toolkit/dictionary_learning
+poetry install
+```
+
+## GPU Support
+
+Make sure you have PyTorch with CUDA support:
+
+```bash
+# Check your CUDA version
+nvcc --version
+
+# Install PyTorch with CUDA (example for CUDA 11.8)
+pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118
+```
+
+## For FairFace Dataset
+
+If you plan to use the FairFace dataset features:
+
+```bash
+# Install dlib (may require system dependencies)
+pip install dlib
+
+# On Ubuntu, if dlib fails, install dependencies first:
+sudo apt-get update
+sudo apt-get install build-essential cmake
+sudo apt-get install libopenblas-dev liblapack-dev
+sudo apt-get install libx11-dev libgtk-3-dev
+pip install dlib
+```
+
+## Verify Installation
+
+```python
+# Test basic imports
+import torch
+print(f"PyTorch version: {torch.__version__}")
+print(f"CUDA available: {torch.cuda.is_available()}")
+
+# Test toolkit imports
+from t2Interp.T2I import T2IModel
+from t2Interp.intervention import EncoderAttentionIntervention
+print("✅ Toolkit imported successfully!")
+
+# Test diffusers
+from diffusers import AutoPipelineForText2Image
+print("✅ Diffusers imported successfully!")
+```
+
+## Common Issues
+
+### Issue: "No module named 't2Interp'"
+**Solution:** Make sure you're in the notebook directory or the toolkit is installed:
+```bash
+cd /home/ubuntu/T2I_Interp_toolkit/notebooks
+# Run your notebook from here
+```
+
+### Issue: dlib installation fails
+**Solution:** Install system dependencies:
+```bash
+sudo apt-get install build-essential cmake libopenblas-dev liblapack-dev
+pip install dlib
+```
+
+### Issue: CUDA out of memory
+**Solution:** Reduce batch size or use mixed precision:
+```python
+# In your code
+pipe = pipe.to("cuda", torch_dtype=torch.float16)
+```
+
+## Environment Setup
+
+For a clean installation, consider using a virtual environment:
+
+```bash
+# Create virtual environment
+python -m venv t2i_env
+
+# Activate it
+source t2i_env/bin/activate
+
+# Install requirements
+pip install -r requirements.txt
+```
+
+## Minimal Requirements
+
+If you just want to run the debiasing notebook:
+
+```bash
+pip install torch torchvision diffusers transformers pandas scikit-learn numpy matplotlib
+```
+
+## Development Installation
+
+For development with all optional dependencies:
+
+```bash
+pip install -r requirements.txt
+pip install pytest black flake8 mypy  # Development tools
+```
+
+