undertherain / benchmarker / 594

46%

master: 44%

push

travis-ci

Fix (mini fix) pytorch output

`batch_idx` is multiplied by `len(data)` (which results in the number
of samples processed). This doesn't match the next value after '/', is
`len(self.x_train)` which is the total number of batches (and not
total number of samples).

This closes #66

98 of 403 branches covered (24.32%)

Branch coverage included in aggregate %.

967 of 1929 relevant lines covered (50.13%)

0.5 hits per line