I try to implement paper association LSTM.
The model is created, but there is a error as follow:
RuntimeError Traceback (most recent call last)
~/experiment/ssd.pytorch-master/train_association_lstm.py in ()
271
272 if name == ‘main’:
→ 273 train()~/experiment/ssd.pytorch-master/train_association_lstm.py in train()
196 loss_l, loss_c = criterion(out, targets)
197 loss = loss_l + loss_c
→ 198 loss.backward()
199 optimizer.step()
200 t1 = time.time()~/anaconda2/envs/gluon/lib/python3.5/site-packages/torch/tensor.py in backward(self, gradient, retain_graph, create_graph)
91 products. Defaults toFalse.
92 “”"
—> 93 torch.autograd.backward(self, gradient, retain_graph, create_graph)
94
95 def register_hook(self, hook):~/anaconda2/envs/gluon/lib/python3.5/site-packages/torch/autograd/init.py in backward(tensors, grad_tensors, retain_graph, create_graph, grad_variables)
87 Variable._execution_engine.run_backward(
88 tensors, grad_tensors, retain_graph, create_graph,
—> 89 allow_unreachable=True) # allow_unreachable flag
90
91RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation
The complete Model code is:
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
from layers import *
from data import vid
import os
from lib.model.roi_pooling.modules.roi_pool import _RoIPooling
from lib.model.roi_crop.modules.roi_crop import _RoICrop
from lib.model.roi_align.modules.roi_align import RoIAlignAvg
from lib.model.utils.net_utils import _affine_grid_gen
from utils.viz import plot_bbox, plot_image
from matplotlib import pyplot as plt
from bnlstm import BNLSTM
from layers.modules import MultiProjectLoss
class association_lstm(nn.Module):
"""Single Shot Multibox Architecture
The network is composed of a base VGG network followed by the
added multibox conv layers. Each multibox layer branches into
1) conv2d for class conf scores
2) conv2d for localization predictions
3) associated priorbox layer to produce default bounding
boxes specific to the layer's feature map size.
See: https://arxiv.org/pdf/1512.02325.pdf for more details.
Args:
phase: (string) Can be "test" or "train"
size: input image size
base: VGG16 layers for input, size of either 300 or 500
extras: extra layers that feed to multibox loc and conf layers
head: "multibox head" consists of loc and conf conv layers
"""
def __init__(self, phase, cfg, size, base, extras, head, num_classes):
super(association_lstm, self).__init__()
self.phase = phase
self.num_classes = num_classes
self.cfg = vid
self.priorbox = PriorBox(self.cfg)
self.priors = Variable(self.priorbox.forward(), volatile=True)
self.size = size
# SSD network
self.vgg = nn.ModuleList(base)
# Layer learns to scale the l2 normalized features from conv4_3
self.L2Norm = L2Norm(512, 20)
self.extras = nn.ModuleList(extras)
self.loc = nn.ModuleList(head[0])
self.conf = nn.ModuleList(head[1])
self.roi_pool = _RoIPooling(self.cfg['POOLING_SIZE'], self.cfg['POOLING_SIZE'], 1.0 / 16.0)
self.roi_align = RoIAlignAvg(self.cfg['POOLING_SIZE'], self.cfg['POOLING_SIZE'], 1.0 / 16.0)
self.grid_size = self.cfg['POOLING_SIZE'] * 2 if self.cfg['CROP_RESIZE_WITH_MAX_POOL'] else self.cfg['POOLING_SIZE']
self.roi_crop = _RoICrop()
self.img_shape = (self.cfg['min_dim'],self.cfg['min_dim'])
self.tensor_len = 4+self.num_classes+49
self.bnlstm1 = BNLSTM(input_size=84, hidden_size=150, batch_first=False, bidirectional=False)
self.bnlstm2 = BNLSTM(input_size=150, hidden_size=300, batch_first=False, bidirectional=False)
self.cls_pred = nn.Linear(300, self.num_classes)
self.bbox_pred = nn.Linear(300, 4)
self.association_pred = nn.Linear(300, 49)
self.MultiProjectLoss = MultiProjectLoss(self.num_classes, 0, True, 3, 0.5 )
if phase == 'vid_train':
self.softmax = nn.Softmax(dim=-1)
#self.detect = Trnsform_target(num_classes, 200, 0.5, 0.01, 0.45)
self.detect = train_target(num_classes, 200, 0.5, 0.01, 0.45)
def forward(self, x, targets):
"""Applies network layers and ops on input image(s) x.
Args:
x: input image or batch of images. Shape: [batch,3,300,300].
Return:
Depending on phase:
test:
Variable(tensor) of output class label predictions,
confidence score, and corresponding location predictions for
each object detected. Shape: [batch,topk,7]
train:
list of concat outputs from:
1: confidence layers, Shape: [batch*num_priors,num_classes]
2: localization layers, Shape: [batch,num_priors*4]
3: priorbox layers, Shape: [2,num_priors*4]
"""
sources = list()
loc = list()
conf = list()
batch_size = x.data.size(0)
print('input image size: ',x.size())
display_img = x[0].clone().cpu().numpy().transpose((1,2,0))
print('display_img size: ',display_img.shape)
# apply vgg up to conv4_3 relu
for k in range(23):
x = self.vgg[k](x)
s = self.L2Norm(x)
sources.append(s)
# apply vgg up to fc7
for k in range(23, len(self.vgg)):
x = self.vgg[k](x)
sources.append(x)
# apply extra layers and cache source layer outputs
for k, v in enumerate(self.extras):
x = F.relu(v(x), inplace=True)
if k % 2 == 1:
sources.append(x)
# apply multibox head to source layers
for (x, l, c) in zip(sources, self.loc, self.conf):
loc.append(l(x).permute(0, 2, 3, 1).contiguous())
conf.append(c(x).permute(0, 2, 3, 1).contiguous())
roi_feat = sources[1]
loc = torch.cat([o.view(o.size(0), -1) for o in loc], 1)
conf = torch.cat([o.view(o.size(0), -1) for o in conf], 1)
if self.phase == "vid_train":
"""
output = self.detect(
loc.view(loc.size(0), -1, 4), # loc preds
self.softmax(conf.view(conf.size(0), -1, self.num_classes)), # conf preds
#conf.view(conf.size(0), -1, self.num_classes),
self.priors.type(type(x.data)), # default boxes
targets
)
"""
output = self.detect(
loc.view(loc.size(0), -1, 4), # loc preds
self.softmax(conf.view(conf.size(0), -1, self.num_classes)), # conf preds
#conf.view(conf.size(0), -1, self.num_classes),
self.priors.type(type(x.data)), # default boxes
)
else:
output = (
loc.view(loc.size(0), -1, 4),
conf.view(conf.size(0), -1, self.num_classes),
self.priors
)
return output # 临时返回
#rois, loc, conf, loc_t, conf_t = output # rois size: batchsize, top_k, 5
rois, loc, conf, priors = output
img_scale = torch.Tensor([self.img_shape[1], self.img_shape[0],
self.img_shape[1], self.img_shape[0]])
rois[:,:,1:] = rois[:,:,1:]*img_scale
if self.cfg['POOLING_MODE'] == 'crop':
grid_xy = _affine_grid_gen(rois.view(-1, 5), roi_feat.size()[2:], self.grid_size)
grid_yx = torch.stack([grid_xy.data[:,:,:,1], grid_xy.data[:,:,:,0]], 3).contiguous()
pooled_feat = self.roi_crop(roi_feat, Variable(grid_yx).detach())
if self.cfg['CROP_RESIZE_WITH_MAX_POOL']:
pooled_feat = F.max_pool2d(pooled_feat, 2, 2)
elif self.cfg['POOLING_MODE'] == 'align':
pooled_feat = self.roi_align(roi_feat, rois.view(-1, 5))
elif self.cfg['POOLING_MODE'] == 'pool':
pooled_feat = self.roi_pool(roi_feat, rois.view(-1,5))
Con1_1 = nn.Conv2d(1024, 1, kernel_size=1, padding=0, dilation=1)
pooled_feat = Con1_1(pooled_feat)
scale = L2Norm(1, 20)
normlize_feat = scale(pooled_feat)
feat = normlize_feat.view(normlize_feat.size(0), normlize_feat.size(1), -1)
feat = feat.squeeze().view(batch_size, feat.size(0)/batch_size, -1)
stacked_tensor = torch.cat((conf,loc,feat),-1)
o1, _ = self.bnlstm1(stacked_tensor)
o2, _ = self.bnlstm2(o1)
cls_pred = self.cls_pred(o2)
bbox_pred = self.bbox_pred(o2)
association_pred = self.association_pred(o2)
return bbox_pred, cls_pred, self.priors
def load_weights(self, base_file):
other, ext = os.path.splitext(base_file)
if ext == '.pkl' or '.pth':
print('Loading weights into state dict...')
self.load_state_dict(torch.load(base_file,
map_location=lambda storage, loc: storage))
print('Finished!')
else:
print('Sorry only .pth and .pkl files supported.')
# This function is derived from torchvision VGG make_layers()
# https://github.com/pytorch/vision/blob/master/torchvision/models/vgg.py
def vgg(cfg, i, batch_norm=False):
layers = []
in_channels = i
for v in cfg:
if v == 'M':
layers += [nn.MaxPool2d(kernel_size=2, stride=2)]
elif v == 'C':
layers += [nn.MaxPool2d(kernel_size=2, stride=2, ceil_mode=True)]
else:
conv2d = nn.Conv2d(in_channels, v, kernel_size=3, padding=1)
if batch_norm:
layers += [conv2d, nn.BatchNorm2d(v), nn.ReLU(inplace=True)]
else:
layers += [conv2d, nn.ReLU(inplace=True)]
in_channels = v
pool5 = nn.MaxPool2d(kernel_size=3, stride=1, padding=1)
conv6 = nn.Conv2d(512, 1024, kernel_size=3, padding=6, dilation=6)
conv7 = nn.Conv2d(1024, 1024, kernel_size=1)
layers += [pool5, conv6,
nn.ReLU(inplace=True), conv7, nn.ReLU(inplace=True)]
return layers
def add_extras(cfg, i, batch_norm=False):
# Extra layers added to VGG for feature scaling
layers = []
in_channels = i
flag = False
for k, v in enumerate(cfg):
if in_channels != 'S':
if v == 'S':
layers += [nn.Conv2d(in_channels, cfg[k + 1],
kernel_size=(1, 3)[flag], stride=2, padding=1)]
else:
layers += [nn.Conv2d(in_channels, v, kernel_size=(1, 3)[flag])]
flag = not flag
in_channels = v
return layers
def multibox(vgg, extra_layers, cfg, num_classes):
loc_layers = []
conf_layers = []
vgg_source = [21, -2]
for k, v in enumerate(vgg_source):
loc_layers += [nn.Conv2d(vgg[v].out_channels,
cfg[k] * 4, kernel_size=3, padding=1)]
conf_layers += [nn.Conv2d(vgg[v].out_channels,
cfg[k] * num_classes, kernel_size=3, padding=1)]
for k, v in enumerate(extra_layers[1::2], 2):
loc_layers += [nn.Conv2d(v.out_channels, cfg[k]
* 4, kernel_size=3, padding=1)]
conf_layers += [nn.Conv2d(v.out_channels, cfg[k]
* num_classes, kernel_size=3, padding=1)]
return vgg, extra_layers, (loc_layers, conf_layers)
base = {
'300': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'C', 512, 512, 512, 'M',
512, 512, 512],
'512': [],
}
extras = {
'300': [256, 'S', 512, 128, 'S', 256, 128, 256, 128, 256],
'512': [],
}
mbox = {
'300': [4, 6, 6, 6, 4, 4], # number of boxes per feature map location
'512': [],
}
def build_association_lstm(phase, cfg, size=300, num_classes=21):
if phase != "test" and phase != "vid_train":
print("ERROR: Phase: " + phase + " not recognized")
return
if size != 300:
print("ERROR: You specified size " + repr(size) + ". However, " +
"currently only SSD300 (size=300) is supported!")
return
base_, extras_, head_ = multibox(vgg(base[str(size)], 3),
add_extras(extras[str(size)], 1024),
mbox[str(size)], num_classes)
return association_lstm(phase, cfg, size, base_, extras_, head_, num_classes)
The loss compute as follow:
class MultiResLoss(nn.Module):
"""SSD Weighted Loss Function
Compute Targets:
1) Produce Confidence Target Indices by matching ground truth boxes
with (default) 'priorboxes' that have jaccard index > threshold parameter
(default threshold: 0.5).
2) Produce localization target by 'encoding' variance into offsets of ground
truth boxes and their matched 'priorboxes'.
3) Hard negative mining to filter the excessive number of negative examples
that comes with using a large number of default bounding boxes.
(default negative:positive ratio 3:1)
Objective Loss:
L(x,c,l,g) = (Lconf(x, c) + αLloc(x,l,g)) / N
Where, Lconf is the CrossEntropy Loss and Lloc is the SmoothL1 Loss
weighted by α which is set to 1 by cross val.
Args:
c: class confidences,
l: predicted boxes,
g: ground truth boxes
N: number of matched default boxes
See: https://arxiv.org/pdf/1512.02325.pdf for more details.
"""
def __init__(self, num_classes, overlap_thresh, prior_for_matching,
bkg_label, neg_mining, neg_pos, neg_overlap, encode_target,
use_gpu=True):
super(MultiResLoss, self).__init__()
self.use_gpu = use_gpu
self.num_classes = num_classes
self.threshold = overlap_thresh
self.background_label = bkg_label
self.encode_target = encode_target
self.use_prior_for_matching = prior_for_matching
self.do_neg_mining = neg_mining
self.negpos_ratio = neg_pos
self.neg_overlap = neg_overlap
self.variance = cfg['variance']
def forward(self, predictions, targets):
"""Multibox Loss
Args:
predictions (tuple): A tuple containing loc preds, conf preds,
and prior boxes from SSD net.
conf shape: torch.size(batch_size,num_priors,num_classes)
loc shape: torch.size(batch_size,num_priors,4)
priors shape: torch.size(num_priors,4)
targets (tensor): Ground truth boxes and labels for a batch,
shape: [batch_size,num_objs,5] (last idx is the label).
"""
loc_data, conf_data, priors = predictions
num = loc_data.size(0)
num_priors = (priors.size(1))
num_classes = self.num_classes
# match priors (default boxes) and ground truth boxes
loc_t = torch.Tensor(num, num_priors, 4)
conf_t = torch.LongTensor(num, num_priors)
for idx in range(num):
truths = targets[idx][:, :-1].data
labels = targets[idx][:, -1].data
#defaults = priors.data
defaults = priors[idx].data
match(self.threshold, truths, defaults, self.variance, labels,
loc_t, conf_t, idx)
if self.use_gpu:
loc_t = loc_t.cuda()
conf_t = conf_t.cuda()
# wrap targets
loc_t = Variable(loc_t, requires_grad=False)
conf_t = Variable(conf_t, requires_grad=False)
print('conf_t dtype: ',conf_t.dtype)
pos = conf_t > 0
num_pos = pos.sum(dim=1, keepdim=True)
pos_idx = pos.unsqueeze(pos.dim()).expand_as(loc_data)
loc_p = loc_data[pos_idx].view(-1, 4)
loc_t = loc_t[pos_idx].view(-1, 4)
loss_l = F.smooth_l1_loss(loc_p, loc_t, size_average=False)
# Compute max conf across batch for hard negative mining
batch_conf = conf_data.view(-1, self.num_classes)
loss_c = log_sum_exp(batch_conf) - batch_conf.gather(1, conf_t.view(-1, 1))
# Hard Negative Mining
# add for pytorch 0.4
loss_c = loss_c.view(pos.size()[0], pos.size()[1])
loss_c[pos] = 0 # filter out pos boxes for now
loss_c = loss_c.view(num, -1)
_, loss_idx = loss_c.sort(1, descending=True)
_, idx_rank = loss_idx.sort(1)
num_pos = pos.long().sum(1, keepdim=True)
num_neg = torch.clamp(self.negpos_ratio*num_pos, max=pos.size(1)-1)
neg = idx_rank < num_neg.expand_as(idx_rank)
# Confidence Loss Including Positive and Negative Examples
pos_idx = pos.unsqueeze(2).expand_as(conf_data)
neg_idx = neg.unsqueeze(2).expand_as(conf_data)
conf_p = conf_data[(pos_idx+neg_idx).gt(0)].view(-1, self.num_classes)
targets_weighted = conf_t[(pos+neg).gt(0)]
loss_c = F.cross_entropy(conf_p, targets_weighted, size_average=False)
# Sum of losses: L(x,c,l,g) = (Lconf(x, c) + αLloc(x,l,g)) / N
N = num_pos.data.sum()
# add for pytorch 0.4
N = num_pos.data.sum().double()
loss_l = loss_l.double()
loss_c = loss_c.double()
###
loss_l /= N
loss_c /= N
return loss_l, loss_c
Hope someone can help me find where the error is.
Thank you very much!