--check-gpu

🦉 AthenaAI is an AI-powered ensemble machine-learning trading bot for PocketOption, built on BinaryOptionsToolsV2 → Get it on GitLab
💬 Need help? Join the Chipa Discord · ⚡ Go no-code with ChipaEditor

Print which device the neural model would train on, then exit. No data is loaded and no training happens.


Type	boolean flag (store_true)
Default	off
Maps to	handled directly in `train.py` `main()` — never reaches `TrainerConfig`
Read in	`train.py` → `src/training/device.py` (`detect_device()`)

What it does

main() checks this flag first. If set, it calls detect_device() (with no arguments, so --cpu is not applied), prints one line, and returns exit code 0:

if args.check_gpu:
    device, desc = detect_device()
    print(f"Device: {device}  ({desc})")
    return 0

detect_device() (src/training/device.py) returns one of:

Output	Meaning
`Device: cuda (GPU: <name> (<N> GB VRAM))`	PyTorch with a working CUDA GPU; name and total VRAM come from `torch.cuda.get_device_name(0)` / device properties
`Device: mps (GPU: Apple Metal (MPS))`	Apple Silicon with MPS available
`Device: cpu (CPU (PyTorch not installed — pip install torch for GPU training))`	torch import failed
`Device: cpu (CPU (PyTorch installed but no GPU available))`	torch present, no CUDA/MPS

All other flags on the command line are ignored when --check-gpu is present.

Choosing a value

Not applicable — it’s a diagnostic switch. Use it:

after installing PyTorch, to confirm the CUDA build is active (if you have an NVIDIA GPU but see cpu, you likely installed the CPU-only wheel — reinstall from the CUDA index, e.g. pip install torch --index-url https://download.pytorch.org/whl/cu121);
before a long training run, to know whether the MLP will use the GPU.

Examples

python train.py --check-gpu
# Device: cuda  (GPU: NVIDIA GeForce RTX 4070 (12.0 GB VRAM))

python train.py --check-gpu

Interactions

--cpu — not consulted by --check-gpu; the check reports what auto-detection finds.
--no-nn — if you plan to skip the NN anyway, the device is irrelevant.

--check-gpu

What it does

Choosing a value

Examples

Interactions

See also