improve CI: use conda as much as possible for MKL optimization etc.

zuoxingdong / lagom

lagom: A PyTorch infrastructure for rapid prototyping of reinforcement learning algorithms.

MIT License

373 stars 30 forks source link

Closed zuoxingdong closed 5 years ago