projects / platform / core / ml / nntrainer.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: efa307b) | patch
[bugfix] Ignoring error in training/validation
authorParichay Kapoor <pk.kapoor@samsung.com>
Fri, 12 Jun 2020 09:24:38 +0000 (18:24 +0900)
committerJijoong Moon <jijoong.moon@samsung.com>
Mon, 15 Jun 2020 00:17:35 +0000 (09:17 +0900)
commitce094be78a73d7e6b8842958d93fad64b50211dc
tree001fd91d2368b170492e229629ae0a50e80e0d91tree | snapshot
parentefa307b44ab32556422c0747658f05c1858d470ccommit | diff
[bugfix] Ignoring error in training/validation

Errors in forward and backward propagation were ignored.
Added error handling for them.

Further, model was being saved after validation.
Errors in validation would result in no model.
Changed to saving of model after training than after validation.

Signed-off-by: Parichay Kapoor <pk.kapoor@samsung.com>
nntrainer/src/neuralnet.cpp diff | blob | history
Domain: Machine Learning / ML Framework;