search

DCASE-REPO / dcase2018_baseline

DCASE 2018 Baseline systems
MIT License
128 stars 78 forks source link

The loss was always 0 when I was training on Task 2 #13

Closed jiangf8686 closed 1 year ago

jiangf8686 commented 6 years ago

The loss was always 0 when I was training on Task 2, and then I had this problem when I verified the effect.

NotFoundError (see above for traceback): Tensor name "Conv/biases" not found in checkpoint files /model.ckpt-14255.index [[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_INT32], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]

FunkMonkey commented 6 years ago

I am experiencing the same on Windows. @jiangf8686 Did you find a solution?