Commit 36b05bb7 authored by Yuxin Wu's avatar Yuxin Wu

Improve warning message on mismatched variables (#901)

parent 2fc3be15
......@@ -13,6 +13,7 @@ It's Yet Another TF high-level API, with __speed__, __readability__ and __flexib
1. Focus on __training speed__.
+ Speed comes for free with Tensorpack -- it uses TensorFlow in the __efficient way__ with no extra overhead.
On common CNNs, it runs training [1.2~5x faster](https://github.com/tensorpack/benchmarks/tree/master/other-wrappers) than the equivalent Keras code.
Your training can probably gets faster if written with Tensorpack.
+ Data-parallel multi-GPU/distributed training strategy is off-the-shelf to use.
It scales as well as Google's [official benchmark](https://www.tensorflow.org/performance/benchmarks).
......
......@@ -140,9 +140,9 @@ class SaverRestore(SessionInit):
func(reader, name, v)
chkpt_vars_used.add(name)
else:
vname = v.op.name
if not is_training_name(vname):
mismatch.add(vname)
# use tensor name (instead of op name) for logging, to be consistent with the reverse case
if not is_training_name(v.name):
mismatch.add(v.name)
mismatch.log()
mismatch = MismatchLogger('checkpoint', 'graph')
if len(chkpt_vars_used) < len(chkpt_vars):
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment