PyBrain does not have built-in support for GPU acceleration. PyBrain is primarily designed for educational and research purposes, and its focus has been on providing a flexible and user-friendly framework for experimenting with various machine learning algorithms and neural network architectures on CPU.

However, if you're interested in leveraging GPU acceleration for neural network training in Python, you may consider using other deep learning libraries that support GPU computation, such as TensorFlow, PyTorch, or Keras (which can run on top of TensorFlow or other backends).

These libraries provide seamless integration with GPUs, allowing you to accelerate the training of deep neural networks significantly compared to CPU-based computations. They offer optimized implementations of neural network operations that take advantage of the parallel processing capabilities of modern GPUs, resulting in faster training times for large-scale datasets and complex models.

If GPU acceleration is a critical requirement for your project, you may need to transition to one of these alternative libraries that offer native GPU support. However, if you prefer to stick with PyBrain for its simplicity and ease of use, you may need to accept the limitation of running computations on CPU only.