Skip to content

Instantly share code, notes, and snippets.

@zmjjmz
Created December 20, 2017 19:05
Show Gist options
  • Save zmjjmz/9edee5b4eeff94f383122545d80ee55f to your computer and use it in GitHub Desktop.
Save zmjjmz/9edee5b4eeff94f383122545d80ee55f to your computer and use it in GitHub Desktop.
tf weight repro no global
Using TensorFlow backend.
2017-12-20 14:04:22.144670: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
output from keras
[[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]]
Export dir exists! Resolving according to conflict policy: bump
Bumping to version 12
{'lookedup': <tf.Tensor 'lookedup:0' shape=(?, 5) dtype=int32>}
{'embed': <tf.Tensor 'embed/Gather:0' shape=(?, 5, 3) dtype=float32>}
model weights
tensor_name: embed/embeddings
[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
tf serving output
[[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]]
keras output
[[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]
[[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]]]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment