Open-source tool estimates neural network cost before GPU runs
Tensey is a visual tool for designing . It checks tensor shapes while the design is being built, counts , and estimates FLOPs and VRAM use.
It can catch broken residual connections and Linear layer size mismatches before GPU time is wasted. It supports 63 and includes shape .
Finished designs can be exported as working PyTorch code. The project is available on GitHub under an .
Key points
- Checks tensor shape problems during design.
- Estimates , FLOPs, and VRAM before running on a GPU.
- Finds residual connection and Linear layer size errors early.
- Supports 63 and exports working PyTorch code.
- Released as an project on GitHub.