Refine lib -> xautodl

CHANGE-LOG.md

@@ -8,4 +8,5 @@
 - [2020.09.16] [7052265] Create NATS-BENCH.
 - [2020.10.15] [446262a](https://github.com/D-X-Y/AutoDL-Projects/tree/446262a) Update NATS-BENCH to version 1.0
 - [2020.12.20] [dae387a](https://github.com/D-X-Y/AutoDL-Projects/tree/dae387a) Update NATS-BENCH to version 1.1
-- [2021.05.18] [98fadf8](https://github.com/D-X-Y/AutoDL-Projects/tree/98fadf8086727f6311d477d0e4f2d5bb7e41ca59) Before moving to `xautodl`
+- [2021.05.18] [98fadf8](https://github.com/D-X-Y/AutoDL-Projects/tree/98fadf8) Before moving to `xautodl`
+- [2021.05.19] [129c9e5](https://github.com/D-X-Y/AutoDL-Projects/tree/129c9e5) `xautodl` is close to ready

docs/CVPR-2019-GDAS.md

@@ -11,7 +11,7 @@ Experiments on CIFAR-10, CIFAR-100, ImageNet, PTB, and WT2 are reported.
 
 ## Requirements and Preparation
 
-Please install `Python>=3.6` and `PyTorch>=1.2.0`.
+Please install `Python>=3.6` and `PyTorch>=1.5.0`.
 
 CIFAR and ImageNet should be downloaded and extracted into `$TORCH_HOME`.
 

xautodl/__init__.py

@@ -7,5 +7,5 @@
 
 
 def version():
-    versions = ["0.9.9"]  # 2021.05.18
+    versions = ["0.9.9"]  # 2021.05.19
     return versions[-1]

xautodl/models/shape_searchs/generic_size_tiny_cell_model.py

@@ -11,7 +11,7 @@ import torch.nn as nn
 
 from ..cell_operations import ResNetBasicblock
 from ..cell_infers.cells import InferCell
-from .shape_searchs.SoftSelect import select2withP, ChannelWiseInter
+from .SoftSelect import select2withP, ChannelWiseInter
 
 
 class GenericNAS301Model(nn.Module):