iicd/xautodl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refine lib -> xautodl

Browse Source
This commit is contained in:
D-X-Y 2021-05-19 07:23:50 +00:00
parent 5b9a028e60
commit bd407ac4dc
3 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/CVPR-2019-GDAS.md
View File

@ -1,6 +1,6 @@
# [Searching for A Robust Neural Architecture in Four GPU Hours](https://arxiv.org/abs/1910.04465) # [Searching for A Robust Neural Architecture in Four GPU Hours](https://arxiv.org/abs/1910.04465)
<img align="right" src="https://d-x-y.github.com/resources/paper-icon/CVPR-2019-GDAS.png" width="300"> <img align="right" src="http://xuanyidong.com/resources/paper-icon/CVPR-2019-GDAS.png" width="300">
Searching for A Robust Neural Architecture in Four GPU Hours is accepted at CVPR 2019. Searching for A Robust Neural Architecture in Four GPU Hours is accepted at CVPR 2019.
In this paper, we proposed a Gradient-based searching algorithm using Differentiable Architecture Sampling (GDAS). In this paper, we proposed a Gradient-based searching algorithm using Differentiable Architecture Sampling (GDAS).

2
docs/ICCV-2019-SETN.md
View File

@ -1,6 +1,6 @@
# [One-Shot Neural Architecture Search via Self-Evaluated Template Network](https://arxiv.org/abs/1910.05733) # [One-Shot Neural Architecture Search via Self-Evaluated Template Network](https://arxiv.org/abs/1910.05733)
<img align="right" src="https://d-x-y.github.com/resources/paper-icon/ICCV-2019-SETN.png" width="450"> <img align="right" src="http://xuanyidong.com/resources/paper-icon/ICCV-2019-SETN.png" width="450">
<strong>Highlight</strong>: we equip one-shot NAS with an architecture sampler and train network weights using uniformly sampling. <strong>Highlight</strong>: we equip one-shot NAS with an architecture sampler and train network weights using uniformly sampling.

4
xautodl/utils/flop_benchmark.py
View File

@ -10,9 +10,9 @@ def count_parameters_in_MB(model):
def count_parameters(model_or_parameters, unit="mb"): def count_parameters(model_or_parameters, unit="mb"):
if isinstance(model_or_parameters, nn.Module): if isinstance(model_or_parameters, nn.Module):
counts = sum(np.prod(v.size()) for v in model_or_parameters.parameters()) counts = sum(np.prod(v.size()) for v in model_or_parameters.parameters())
elif isinstance(models_or_parameters, nn.Parameter): elif isinstance(model_or_parameters, nn.Parameter):
counts = models_or_parameters.numel() counts = models_or_parameters.numel()
elif isinstance(models_or_parameters, (list, tuple)): elif isinstance(model_or_parameters, (list, tuple)):
counts = sum(count_parameters(x, None) for x in models_or_parameters) counts = sum(count_parameters(x, None) for x in models_or_parameters)
else: else:
counts = sum(np.prod(v.size()) for v in model_or_parameters) counts = sum(np.prod(v.size()) for v in model_or_parameters)