Closed schwittlick closed 7 years ago
and check logs for models with smallest loss result.
/projects/eco/src/python/keras_lstm/logs/arthur_danto-the_end_of_art_a_philosophical_defence_i10e10s2.log
shows iteration 3 has lowest loss of 0.8
loss of iteration 10 is 2.4 (worse than iteration 3)
i dont really know what the loss is. maybe gotta read this http://deeplearning4j.org/lstm.html
and check logs for models with smallest loss result.
shows iteration 3 has lowest loss of 0.8
loss of iteration 10 is 2.4 (worse than iteration 3)
i dont really know what the loss is. maybe gotta read this http://deeplearning4j.org/lstm.html