Add v4rec hgnet (PaddlePaddle#9768)

* v4rec code

* v4rec add nrtrloss

* Add V4rec backbone file

* Add V4Rec config file.

* Fix V4rec reparameters when export_model

* convert lvnetv3

* fix codestyle

* fix infer_rec v4rec

* add v4rec hgnet

* add v4rec hgnet config

* add svtr_hgnet

* fix bugs in infer_rec and hgnet

configs/rec/PP-OCRv4/ch_PP-OCRv4_rec_hgnet.yml

@@ -0,0 +1,131 @@
+Global:
+ debug: false
+ use_gpu: true
+ epoch_num: 200
+ log_smooth_window: 20
+ print_batch_step: 10
+ save_model_dir: ./output/rec_ppocr_v4_hgnet
+ save_epoch_step: 10
+ eval_batch_step: [0, 2000]
+ cal_metric_during_train: true
+ pretrained_model:
+ checkpoints:
+ save_inference_dir:
+ use_visualdl: false
+ infer_img: doc/imgs_words/ch/word_1.jpg
+ character_dict_path: ppocr/utils/ppocr_keys_v1.txt
+ max_text_length: &max_text_length 25
+ infer_mode: false
+ use_space_char: true
+ distributed: true
+ save_res_path: ./output/rec/predicts_ppocrv3.txt
+
+
+Optimizer:
+ name: Adam
+ beta1: 0.9
+ beta2: 0.999
+ lr:
+ name: Cosine
+ learning_rate: 0.001
+ warmup_epoch: 5
+ regularizer:
+ name: L2
+ factor: 3.0e-05
+
+
+Architecture:
+ model_type: rec
+ algorithm: SVTR_HGNet
+ Transform:
+ Backbone:
+ name: PPHGNet_small
+ Head:
+ name: MultiHead
+ head_list:
+ - CTCHead:
+ Neck:
+ name: svtr
+ dims: 120
+ depth: 2
+ hidden_dims: 120
+ kernel_size: [1, 3]
+ use_guide: True
+ Head:
+ fc_decay: 0.00001
+ - NRTRHead:
+ nrtr_dim: 384
+ max_text_length: *max_text_length
+
+Loss:
+ name: MultiLoss
+ loss_config_list:
+ - CTCLoss:
+ - NRTRLoss:
+
+PostProcess: 
+ name: CTCLabelDecode
+
+Metric:
+ name: RecMetric
+ main_indicator: acc
+
+Train:
+ dataset:
+ name: SimpleDataSet
+ data_dir: ./train_data/
+ ext_op_transform_idx: 1
+ label_file_list:
+ - ./train_data/train_list.txt
+ transforms:
+ - DecodeImage:
+ img_mode: BGR
+ channel_first: false
+ - RecConAug:
+ prob: 0.5
+ ext_data_num: 2
+ image_shape: [48, 320, 3]
+ max_text_length: *max_text_length
+ - RecAug:
+ - MultiLabelEncode:
+ gtc_encode: NRTRLabelEncode
+ - RecResizeImg:
+ image_shape: [3, 48, 320]
+ - KeepKeys:
+ keep_keys:
+ - image
+ - label_ctc
+ - label_gtc
+ - length
+ - valid_ratio
+ loader:
+ shuffle: true
+ batch_size_per_card: 128
+ drop_last: true
+ num_workers: 4
+Eval:
+ dataset:
+ name: SimpleDataSet
+ data_dir: ./train_data
+ label_file_list:
+ - ./train_data/val_list.txt
+ transforms:
+ - DecodeImage:
+ img_mode: BGR
+ channel_first: false
+ - MultiLabelEncode:
+ gtc_encode: NRTRLabelEncode
+ - RecResizeImg:
+ image_shape: [3, 48, 320]
+ - KeepKeys:
+ keep_keys:
+ - image
+ - label_ctc
+ - label_gtc
+ - length
+ - valid_ratio
+ loader:
+ shuffle: false
+ drop_last: false
+ batch_size_per_card: 128
+ num_workers: 4

ppocr/modeling/backbones/__init__.py

@@ -46,11 +46,12 @@ def build_backbone(config, model_type):
  from .rec_densenet import DenseNet
  from .rec_shallow_cnn import ShallowCNN
  from .rec_lcnetv3 import LCNetv3
+ from .rec_hgnet import PPHGNet_small
  support_dict = [
  'MobileNetV1Enhance', 'MobileNetV3', 'ResNet', 'ResNetFPN', 'MTB',
  'ResNet31', 'ResNet45', 'ResNet_ASTER', 'MicroNet',
  'EfficientNetb3_PREN', 'SVTRNet', 'ViTSTR', 'ResNet32', 'ResNetRFL',
- 'DenseNet', 'ShallowCNN', 'LCNetv3'
+ 'DenseNet', 'ShallowCNN', 'LCNetv3', 'PPHGNet_small'
  ]
  elif model_type == 'e2e':
  from .e2e_resnet_vd_pg import ResNet