Popular repositories Loading
-
-
True-Implementation-of-Mixed-Precision-Quantization
True-Implementation-of-Mixed-Precision-Quantization PublicThis code aims to reduce storage or communication bit-width through true quantization. Existing methods quantize gradients (e.g., 2-bit) but store data in torch.float32 or torch.float16, making it …
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.