def test_cost_layer(self):
cost1 = layer.classification_cost(input=inference, label=label)
cost2 = layer.classification_cost(input=inference,
label=label,
weight=weight)
cost3 = layer.cross_entropy_cost(input=inference, label=label)
cost4 = layer.cross_entropy_with_selfnorm_cost(input=inference,
label=label)
cost5 = layer.mse_cost(input=inference, label=label)
cost6 = layer.mse_cost(input=inference, label=label, weight=weight)
cost7 = layer.multi_binary_label_cross_entropy_cost(input=inference,
label=label)
cost8 = layer.rank_cost(left=score, right=score, label=score)
cost9 = layer.lambda_cost(input=inference, score=score)
cost10 = layer.sum_cost(input=inference)
cost11 = layer.huber_cost(input=score, label=label)
print layer.parse_network(cost1, cost2)
print layer.parse_network(cost3, cost4)
print layer.parse_network(cost5, cost6)
print layer.parse_network(cost7, cost8, cost9, cost10, cost11)
crf = layer.crf(input=inference, label=label)
crf_decoding = layer.crf_decoding(input=inference, size=3)
ctc = layer.ctc(input=inference, label=label)
warp_ctc = layer.warp_ctc(input=pixel, label=label)
nce = layer.nce(input=inference, label=label, num_classes=3)
hsigmoid = layer.hsigmoid(input=inference, label=label, num_classes=3)
print layer.parse_network(crf, crf_decoding, ctc, warp_ctc, nce,
hsigmoid)
def test_cost_layer(self):
cost1 = layer.classification_cost(input=inference, label=label)
cost2 = layer.classification_cost(
input=inference, label=label, weight=weight)
cost3 = layer.cross_entropy_cost(input=inference, label=label)
cost4 = layer.cross_entropy_with_selfnorm_cost(
input=inference, label=label)
cost5 = layer.square_error_cost(input=inference, label=label)
cost6 = layer.square_error_cost(
input=inference, label=label, weight=weight)
cost7 = layer.multi_binary_label_cross_entropy_cost(
input=inference, label=label)
cost8 = layer.rank_cost(left=score, right=score, label=score)
cost9 = layer.lambda_cost(input=inference, score=score)
cost10 = layer.sum_cost(input=inference)
cost11 = layer.huber_regression_cost(input=score, label=label)
cost12 = layer.huber_classification_cost(input=score, label=label)
print layer.parse_network([cost1, cost2])
print layer.parse_network([cost3, cost4])
print layer.parse_network([cost5, cost6])
print layer.parse_network([cost7, cost8, cost9, cost10, cost11, cost12])
crf = layer.crf(input=inference, label=label)
crf_decoding = layer.crf_decoding(input=inference, size=3)
ctc = layer.ctc(input=inference, label=label)
warp_ctc = layer.warp_ctc(input=pixel, label=label)
nce = layer.nce(input=inference, label=label, num_classes=3)
hsigmoid = layer.hsigmoid(input=inference, label=label, num_classes=3)
print layer.parse_network(
[crf, crf_decoding, ctc, warp_ctc, nce, hsigmoid])
def __build_nn__(self):
'''
Build the network topology.
'''
# Get the image features with CNN.
conv_features = self.conv_groups(self.image, conf.filter_num,
conf.with_bn)
# Expand the output of CNN into a sequence of feature vectors.
sliced_feature = layer.block_expand(
input=conv_features,
num_channels=conf.num_channels,
stride_x=conf.stride_x,
stride_y=conf.stride_y,
block_x=conf.block_x,
block_y=conf.block_y)
# Use RNN to capture sequence information forwards and backwards.
gru_forward = simple_gru(
input=sliced_feature, size=conf.hidden_size, act=Relu())
gru_backward = simple_gru(
input=sliced_feature,
size=conf.hidden_size,
act=Relu(),
reverse=True)
# Map the output of RNN to character distribution.
self.output = layer.fc(input=[gru_forward, gru_backward],
size=self.num_classes + 1,
act=Linear())
self.log_probs = paddle.layer.mixed(
input=paddle.layer.identity_projection(input=self.output),
act=paddle.activation.Softmax())
# Use warp CTC to calculate cost for a CTC task.
if not self.is_infer:
self.cost = layer.warp_ctc(
input=self.output,
label=self.label,
size=self.num_classes + 1,
norm_by_times=conf.norm_by_times,
blank=self.num_classes)
self.eval = evaluator.ctc_error(input=self.output, label=self.label)
def __build_nn__(self):
'''
建立网络拓扑
'''
# 通过CNN获取图像特征
conv_features = self.conv_groups(self.image, conf.filter_num,
conf.with_bn)
# 将CNN的输出展开成一系列特征向量。
sliced_feature = layer.block_expand(
input=conv_features,
num_channels=conf.num_channels,
stride_x=conf.stride_x,
stride_y=conf.stride_y,
block_x=conf.block_x,
block_y=conf.block_y)
# 使用RNN向前和向后捕获序列信息。
gru_forward = simple_gru(
input=sliced_feature, size=conf.hidden_size, act=Relu())
gru_backward = simple_gru(
input=sliced_feature,
size=conf.hidden_size,
act=Relu(),
reverse=True)
# 将RNN的输出映射到字符分布。
self.output = layer.fc(input=[gru_forward, gru_backward],
size=self.num_classes + 1,
act=Linear())
self.log_probs = paddle.layer.mixed(
input=paddle.layer.identity_projection(input=self.output),
act=paddle.activation.Softmax())
# 使用扭曲CTC来计算CTC任务的成本。
if not self.is_infer:
self.cost = layer.warp_ctc(
input=self.output,
label=self.label,
size=self.num_classes + 1,
norm_by_times=conf.norm_by_times,
blank=self.num_classes)
self.eval = evaluator.ctc_error(input=self.output, label=self.label)