Commit 03b92aba authored by Yuxin Wu's avatar Yuxin Wu

update cifar number with new bn

parent 5a8d500c
......@@ -13,15 +13,15 @@ from tensorpack.tfutils.symbolic_functions import *
from tensorpack.tfutils.summary import *
"""
CIFAR10-resnet example.
CIFAR10 ResNet example. See:
Deep Residual Learning for Image Recognition, arxiv:1512.03385
This implementation uses the variants proposed in:
Identity Mappings in Deep Residual Networks, arxiv:1603.05027
I can reproduce the results for
n=5, about 7.1% val error after 67k step with 2 TitanX (6.1it/s)
n=18, about 6.0% val error after 62k step with 2 TitanX (about 10hr)
n=30: a 182-layer network, about 5.6% val error after 51k step with 2 GPUs
I can reproduce the results on 2 TitanX for
n=5, about 7.1% val error after 67k step (8.6 step/s)
n=18, about 5.7% val error (2.45 step/s)
n=30: a 182-layer network, about 5.6% val error after 51k step (1.55 step/s)
This model uses the whole training set instead of a train-val split.
"""
......@@ -169,7 +169,7 @@ def get_config():
session_config=sess_config,
model=Model(n=18),
step_per_epoch=step_per_epoch,
max_epoch=500,
max_epoch=400,
)
if __name__ == '__main__':
......
......@@ -13,9 +13,14 @@ import tensorpack.tfutils.symbolic_functions as symbf
from tensorpack.tfutils.summary import *
"""
A small convnet model for cifar 10 or cifar100 dataset.
A small convnet model for Cifar10 or Cifar100 dataset.
For Cifar10: 90% validation accuracy after 40k step.
Cifar10:
90% validation accuracy after 40k step.
91% accuracy after 80k step.
18.8 step/s on TitanX
Not a good for Cifar100, just for demonstration.
"""
class Model(ModelDesc):
......
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