Installation

Installation and Environments

Modern Qwen LoRA (Python 3.12, Windows PowerShell)

py -3.12 -m venv .venv-qwen
.\.venv-qwen\Scripts\python.exe -m pip install --upgrade pip
.\.venv-qwen\Scripts\python.exe -m pip install -r requirements.txt
.\.venv-qwen\Scripts\python.exe -m pip check
.\.venv-qwen\Scripts\python.exe qwen_alien_romulus_finetune.py --validate-data

Legacy TRL experiment (new venv; do not reuse Qwen env)

py -3.12 -m venv .venv-legacy
.\.venv-legacy\Scripts\python.exe -m pip install --upgrade pip
.\.venv-legacy\Scripts\python.exe -m pip install -r requirements_legacy_llm_v2.txt
.\.venv-legacy\Scripts\python.exe LLM_Training_Fixed_v2.py --mode validate
.\.venv-legacy\Scripts\python.exe LLM_Training_Fixed_v2.py --mode check

Compute and compatibility

• Model download requires network access and potentially a Hugging Face token for gated checkpoints.
• NVIDIA GPU training requires a CUDA-enabled PyTorch build compatible with local hardware; choose from the official PyTorch installer.
• A 4 GB VRAM GPU may run out of memory even with LoRA. CPU training can be very slow.
• TRL 0.8.6 has version-pinned API requirements and can conflict with modern Colab images; keep it isolated.
• Notebook users should select matching virtual environment kernels (install ipykernel in each).

No install is required for the bare Python validation modes besides a suitable Python interpreter.