LDOUBLEV
4 years ago
commit
56cbbdfb01
@ -0,0 +1,107 @@
|
||||
Global:
|
||||
use_gpu: True
|
||||
epoch_num: 72
|
||||
log_smooth_window: 20
|
||||
print_batch_step: 5
|
||||
save_model_dir: ./output/rec/srn_new
|
||||
save_epoch_step: 3
|
||||
# evaluation is run every 5000 iterations after the 4000th iteration
|
||||
eval_batch_step: [0, 5000]
|
||||
# if pretrained_model is saved in static mode, load_static_weights must set to True
|
||||
cal_metric_during_train: True
|
||||
pretrained_model:
|
||||
checkpoints:
|
||||
save_inference_dir:
|
||||
use_visualdl: False
|
||||
infer_img: doc/imgs_words/ch/word_1.jpg
|
||||
# for data or label process
|
||||
character_dict_path:
|
||||
character_type: en
|
||||
max_text_length: 25
|
||||
num_heads: 8
|
||||
infer_mode: False
|
||||
use_space_char: False
|
||||
|
||||
|
||||
Optimizer:
|
||||
name: Adam
|
||||
beta1: 0.9
|
||||
beta2: 0.999
|
||||
clip_norm: 10.0
|
||||
lr:
|
||||
learning_rate: 0.0001
|
||||
|
||||
Architecture:
|
||||
model_type: rec
|
||||
algorithm: SRN
|
||||
in_channels: 1
|
||||
Transform:
|
||||
Backbone:
|
||||
name: ResNetFPN
|
||||
Head:
|
||||
name: SRNHead
|
||||
max_text_length: 25
|
||||
num_heads: 8
|
||||
num_encoder_TUs: 2
|
||||
num_decoder_TUs: 4
|
||||
hidden_dims: 512
|
||||
|
||||
Loss:
|
||||
name: SRNLoss
|
||||
|
||||
PostProcess:
|
||||
name: SRNLabelDecode
|
||||
|
||||
Metric:
|
||||
name: RecMetric
|
||||
main_indicator: acc
|
||||
|
||||
Train:
|
||||
dataset:
|
||||
name: LMDBDataSet
|
||||
data_dir: ./train_data/srn_train_data_duiqi
|
||||
transforms:
|
||||
- DecodeImage: # load image
|
||||
img_mode: BGR
|
||||
channel_first: False
|
||||
- SRNLabelEncode: # Class handling label
|
||||
- SRNRecResizeImg:
|
||||
image_shape: [1, 64, 256]
|
||||
- KeepKeys:
|
||||
keep_keys: ['image',
|
||||
'label',
|
||||
'length',
|
||||
'encoder_word_pos',
|
||||
'gsrm_word_pos',
|
||||
'gsrm_slf_attn_bias1',
|
||||
'gsrm_slf_attn_bias2'] # dataloader will return list in this order
|
||||
loader:
|
||||
shuffle: False
|
||||
batch_size_per_card: 64
|
||||
drop_last: False
|
||||
num_workers: 4
|
||||
|
||||
Eval:
|
||||
dataset:
|
||||
name: LMDBDataSet
|
||||
data_dir: ./train_data/data_lmdb_release/evaluation
|
||||
transforms:
|
||||
- DecodeImage: # load image
|
||||
img_mode: BGR
|
||||
channel_first: False
|
||||
- SRNLabelEncode: # Class handling label
|
||||
- SRNRecResizeImg:
|
||||
image_shape: [1, 64, 256]
|
||||
- KeepKeys:
|
||||
keep_keys: ['image',
|
||||
'label',
|
||||
'length',
|
||||
'encoder_word_pos',
|
||||
'gsrm_word_pos',
|
||||
'gsrm_slf_attn_bias1',
|
||||
'gsrm_slf_attn_bias2']
|
||||
loader:
|
||||
shuffle: False
|
||||
drop_last: False
|
||||
batch_size_per_card: 32
|
||||
num_workers: 4
|
||||
@ -0,0 +1,47 @@
|
||||
# copyright (c) 2020 PaddlePaddle Authors. All Rights Reserve.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
from __future__ import absolute_import
|
||||
from __future__ import division
|
||||
from __future__ import print_function
|
||||
|
||||
import paddle
|
||||
from paddle import nn
|
||||
|
||||
|
||||
class SRNLoss(nn.Layer):
|
||||
def __init__(self, **kwargs):
|
||||
super(SRNLoss, self).__init__()
|
||||
self.loss_func = paddle.nn.loss.CrossEntropyLoss(reduction="sum")
|
||||
|
||||
def forward(self, predicts, batch):
|
||||
predict = predicts['predict']
|
||||
word_predict = predicts['word_out']
|
||||
gsrm_predict = predicts['gsrm_out']
|
||||
label = batch[1]
|
||||
|
||||
casted_label = paddle.cast(x=label, dtype='int64')
|
||||
casted_label = paddle.reshape(x=casted_label, shape=[-1, 1])
|
||||
|
||||
cost_word = self.loss_func(word_predict, label=casted_label)
|
||||
cost_gsrm = self.loss_func(gsrm_predict, label=casted_label)
|
||||
cost_vsfd = self.loss_func(predict, label=casted_label)
|
||||
|
||||
cost_word = paddle.reshape(x=paddle.sum(cost_word), shape=[1])
|
||||
cost_gsrm = paddle.reshape(x=paddle.sum(cost_gsrm), shape=[1])
|
||||
cost_vsfd = paddle.reshape(x=paddle.sum(cost_vsfd), shape=[1])
|
||||
|
||||
sum_cost = cost_word * 3.0 + cost_vsfd + cost_gsrm * 0.15
|
||||
|
||||
return {'loss': sum_cost, 'word_loss': cost_word, 'img_loss': cost_vsfd}
|
||||
File diff suppressed because it is too large
Load Diff
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in new issue