* fix bug in distributed loss test and change some config to pass at both 2&8 gpus * fix doc
* enable trainer test cases on xpu Signed-off-by: Matrix Yao <matrix.yao@intel.com> * fix style Signed-off-by: Matrix Yao <matrix.yao@intel.com> --------- Signed-off-by: Matrix Yao <matrix.yao@intel.com>
* Fix multi gpu loss sync condition, add doc and test * rename function and class * loss should not scale during inference * fix typo