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

simplify DARTS codes and update affine/track

Browse Source
This commit is contained in:
D-X-Y
2020-01-11 18:46:31 +11:00
parent c66afa4df8
commit 654015bf9d
15 changed files with 30 additions and 110 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
lib/models/cell_searchs/__init__.py
View File

@@ -1,16 +1,15 @@
##################################################
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2019 #
##################################################
from .search_model_darts_v1 import TinyNetworkDartsV1
from .search_model_darts_v2 import TinyNetworkDartsV2
from .search_model_darts import TinyNetworkDarts
from .search_model_gdas import TinyNetworkGDAS
from .search_model_setn import TinyNetworkSETN
from .search_model_enas import TinyNetworkENAS
from .search_model_random import TinyNetworkRANDOM
from .genotypes import Structure as CellStructure, architectures as CellArchitectures
nas_super_nets = {'DARTS-V1': TinyNetworkDartsV1,
'DARTS-V2': TinyNetworkDartsV2,
nas_super_nets = {'DARTS-V1': TinyNetworkDarts,
'DARTS-V2': TinyNetworkDarts,
'GDAS' : TinyNetworkGDAS,
'SETN' : TinyNetworkSETN,
'ENAS' : TinyNetworkENAS,

8
lib/models/cell_searchs/search_model_darts_v1.py → lib/models/cell_searchs/search_model_darts.py
View File

@@ -11,10 +11,10 @@ from .search_cells import SearchCell
from .genotypes import Structure
class TinyNetworkDartsV1(nn.Module):
class TinyNetworkDarts(nn.Module):
def __init__(self, C, N, max_nodes, num_classes, search_space):
super(TinyNetworkDartsV1, self).__init__()
def __init__(self, C, N, max_nodes, num_classes, search_space, affine, track_running_stats):
super(TinyNetworkDarts, self).__init__()
self._C = C
self._layerN = N
self.max_nodes = max_nodes
@@ -31,7 +31,7 @@ class TinyNetworkDartsV1(nn.Module):
if reduction:
cell = ResNetBasicblock(C_prev, C_curr, 2)
else:
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space)
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space, affine, track_running_stats)
if num_edge is None: num_edge, edge2index = cell.num_edges, cell.edge2index
else: assert num_edge == cell.num_edges and edge2index == cell.edge2index, 'invalid {:} vs. {:}.'.format(num_edge, cell.num_edges)
self.cells.append( cell )

93
lib/models/cell_searchs/search_model_darts_v2.py
View File

@@ -1,93 +0,0 @@
##################################################
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2019 #
########################################################
# DARTS: Differentiable Architecture Search, ICLR 2019 #
########################################################
import torch
import torch.nn as nn
from copy import deepcopy
from ..cell_operations import ResNetBasicblock
from .search_cells import SearchCell
from .genotypes import Structure
class TinyNetworkDartsV2(nn.Module):
def __init__(self, C, N, max_nodes, num_classes, search_space):
super(TinyNetworkDartsV2, self).__init__()
self._C = C
self._layerN = N
self.max_nodes = max_nodes
self.stem = nn.Sequential(
nn.Conv2d(3, C, kernel_size=3, padding=1, bias=False),
nn.BatchNorm2d(C))
layer_channels = [C ] * N + [C*2 ] + [C*2 ] * N + [C*4 ] + [C*4 ] * N
layer_reductions = [False] * N + [True] + [False] * N + [True] + [False] * N
C_prev, num_edge, edge2index = C, None, None
self.cells = nn.ModuleList()
for index, (C_curr, reduction) in enumerate(zip(layer_channels, layer_reductions)):
if reduction:
cell = ResNetBasicblock(C_prev, C_curr, 2)
else:
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space)
if num_edge is None: num_edge, edge2index = cell.num_edges, cell.edge2index
else: assert num_edge == cell.num_edges and edge2index == cell.edge2index, 'invalid {:} vs. {:}.'.format(num_edge, cell.num_edges)
self.cells.append( cell )
C_prev = cell.out_dim
self.op_names = deepcopy( search_space )
self._Layer = len(self.cells)
self.edge2index = edge2index
self.lastact = nn.Sequential(nn.BatchNorm2d(C_prev), nn.ReLU(inplace=True))
self.global_pooling = nn.AdaptiveAvgPool2d(1)
self.classifier = nn.Linear(C_prev, num_classes)
self.arch_parameters = nn.Parameter( 1e-3*torch.randn(num_edge, len(search_space)) )
def get_weights(self):
xlist = list( self.stem.parameters() ) + list( self.cells.parameters() )
xlist+= list( self.lastact.parameters() ) + list( self.global_pooling.parameters() )
xlist+= list( self.classifier.parameters() )
return xlist
def get_alphas(self):
return [self.arch_parameters]
def get_message(self):
string = self.extra_repr()
for i, cell in enumerate(self.cells):
string += '\n {:02d}/{:02d} :: {:}'.format(i, len(self.cells), cell.extra_repr())
return string
def extra_repr(self):
return ('{name}(C={_C}, Max-Nodes={max_nodes}, N={_layerN}, L={_Layer})'.format(name=self.__class__.__name__, **self.__dict__))
def genotype(self):
genotypes = []
for i in range(1, self.max_nodes):
xlist = []
for j in range(i):
node_str = '{:}<-{:}'.format(i, j)
with torch.no_grad():
weights = self.arch_parameters[ self.edge2index[node_str] ]
op_name = self.op_names[ weights.argmax().item() ]
xlist.append((op_name, j))
genotypes.append( tuple(xlist) )
return Structure( genotypes )
def forward(self, inputs):
alphas = nn.functional.softmax(self.arch_parameters, dim=-1)
feature = self.stem(inputs)
for i, cell in enumerate(self.cells):
if isinstance(cell, SearchCell):
feature = cell(feature, alphas)
else:
feature = cell(feature)
out = self.lastact(feature)
out = self.global_pooling( out )
out = out.view(out.size(0), -1)
logits = self.classifier(out)
return out, logits

4
lib/models/cell_searchs/search_model_enas.py
View File

@@ -14,7 +14,7 @@ from .search_model_enas_utils import Controller
class TinyNetworkENAS(nn.Module):
def __init__(self, C, N, max_nodes, num_classes, search_space):
def __init__(self, C, N, max_nodes, num_classes, search_space, affine, track_running_stats):
super(TinyNetworkENAS, self).__init__()
self._C = C
self._layerN = N
@@ -32,7 +32,7 @@ class TinyNetworkENAS(nn.Module):
if reduction:
cell = ResNetBasicblock(C_prev, C_curr, 2)
else:
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space)
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space, affine, track_running_stats)
if num_edge is None: num_edge, edge2index = cell.num_edges, cell.edge2index
else: assert num_edge == cell.num_edges and edge2index == cell.edge2index, 'invalid {:} vs. {:}.'.format(num_edge, cell.num_edges)
self.cells.append( cell )

4
lib/models/cell_searchs/search_model_random.py
View File

@@ -13,7 +13,7 @@ from .genotypes import Structure
class TinyNetworkRANDOM(nn.Module):
def __init__(self, C, N, max_nodes, num_classes, search_space):
def __init__(self, C, N, max_nodes, num_classes, search_space, affine, track_running_stats):
super(TinyNetworkRANDOM, self).__init__()
self._C = C
self._layerN = N
@@ -31,7 +31,7 @@ class TinyNetworkRANDOM(nn.Module):
if reduction:
cell = ResNetBasicblock(C_prev, C_curr, 2)
else:
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space)
cell = SearchCell(C_prev, C_curr, 1, max_nodes, search_space, affine, track_running_stats)
if num_edge is None: num_edge, edge2index = cell.num_edges, cell.edge2index
else: assert num_edge == cell.num_edges and edge2index == cell.edge2index, 'invalid {:} vs. {:}.'.format(num_edge, cell.num_edges)
self.cells.append( cell )