添加tps网络

configs/rec/rec_r34_vd_tps_bilstm_ctc.yml

@@ -0,0 +1,100 @@
+Global:
+  use_gpu: true
+  epoch_num: 72
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/r34_vd_tps_bilstm_ctc/
+  save_epoch_step: 3
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [0, 2000]
+  # 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
+  infer_mode: False
+  use_space_char: False
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    learning_rate: 0.0005
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+Architecture:
+  model_type: rec
+  algorithm: CRNN
+  Transform:
+    name: TPS
+    num_fiducial: 20
+    loc_lr: 0.1
+    model_name: small
+  Backbone:
+    name: ResNet
+    layers: 34
+  Neck:
+    name: SequenceEncoder
+    encoder_type: rnn
+    hidden_size: 256
+  Head:
+    name: CTCHead
+    fc_decay: 0
+
+Loss:
+  name: CTCLoss
+
+PostProcess:
+  name: CTCLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: LMDBDateSet
+    data_dir: ./train_data/data_lmdb_release/training/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - CTCLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [3, 32, 100]
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 256
+    drop_last: True
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: LMDBDateSet
+    data_dir: ./train_data/data_lmdb_release/validation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - CTCLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [3, 32, 100]
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 256
+    num_workers: 4

ppocr/modeling/architectures/base_model.py

@@ -16,13 +16,14 @@ from __future__ import division
 from __future__ import print_function
 
 from paddle import nn
-
+from ppocr.modeling.transform import build_transform
 from ppocr.modeling.backbones import build_backbone
 from ppocr.modeling.necks import build_neck
 from ppocr.modeling.heads import build_head
 
 __all__ = ['BaseModel']
 
+
 class BaseModel(nn.Layer):
     def __init__(self, config):
         """
@@ -31,7 +32,7 @@ class BaseModel(nn.Layer):
             config (dict): the super parameters for module.
         """
         super(BaseModel, self).__init__()
-        
+
         in_channels = config.get('in_channels', 3)
         model_type = config['model_type']
         # build transfrom,
@@ -50,7 +51,7 @@ class BaseModel(nn.Layer):
         config["Backbone"]['in_channels'] = in_channels
         self.backbone = build_backbone(config["Backbone"], model_type)
         in_channels = self.backbone.out_channels
-        
+
         # build neck
         # for rec, neck can be cnn,rnn or reshape(None)
         # for det, neck can be FPN, BIFPN and so on.
@@ -62,7 +63,7 @@ class BaseModel(nn.Layer):
             config['Neck']['in_channels'] = in_channels
             self.neck = build_neck(config['Neck'])
             in_channels = self.neck.out_channels
-        
+
         # # build head, head is need for det, rec and cls
         config["Head"]['in_channels'] = in_channels
         self.head = build_head(config["Head"])
@@ -74,4 +75,4 @@ class BaseModel(nn.Layer):
         if self.use_neck:
             x = self.neck(x)
         x = self.head(x)
-        return x
+        return x

ppocr/modeling/transform/__init__.py

@@ -16,7 +16,9 @@ __all__ = ['build_transform']
 
 
 def build_transform(config):
-    support_dict = ['']
+    from .tps import TPS
+
+    support_dict = ['TPS']
 
     module_name = config.pop('name')
     assert module_name in support_dict, Exception(