Thiago Crepaldi aabed34d5c Fix checkpoint API and improve loss scaler handling (#4950) ... This PR also includes: * More LossScaler tests * Minor LossScaler improvement * Check model after extra post processing * Improve basic training tests to include all optimizers * Set rtol=1e-7 tolerance for Legacy vs Experimental frontend API tests * Increase number of training tests for Legacy vs Experimental tests * Minor refactoring on existing tests * Fix Checkpoint API for Gradient Accumulation / fp16 scenarios		2020-09-02 09:38:02 -07:00
..
orttraining	Fix checkpoint API and improve loss scaler handling (#4950)	2020-09-02 09:38:02 -07:00
pytorch_frontend_examples	Fix mnist example (#4926)	2020-08-26 15:28:39 -07:00
tools	Update convergence baseline for ci_test. (#4465)	2020-07-09 15:29:36 +08:00