Closed DesmondYuan closed 3 years ago
A few notes:
expr_name
to name subfolders under figs
and checkpoints
Seems like we have all the features we need #3 #4 #5. Shall we now do some quick test of the default config?
I think we are good to go, to get some sense of
Conclusion1: use a smaller lr, e.g. 1f-3
issue: #6
Are those results from endpoint data instead of dynamic data? We may start from dynamic data which should be much easier
Are those results from endpoint data instead of dynamic data? We may start from dynamic data which should be much easier
Dynamic for test01 and test02. ntotal=1 only for test03
For the initial benchmarking, use
Adding an inference loss in addition to the prediction loss #4 implemented at 09cc46afffc704b96a08ffa1a315517a37b9f499 0c00a66b74e5b6620aa2c9d61d2ef3bf65c09bc3