first commit

configs/rec/rec_vitstr_none_ce.yml

@@ -0,0 +1,102 @@
+Global:
+  use_gpu: True
+  epoch_num: 20
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/vitstr_none_ce/
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations after the 0th iteration#
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words_en/word_10.png
+  # for data or label process
+  character_dict_path: ppocr/utils/EN_symbol_dict.txt
+  max_text_length: 25
+  infer_mode: False
+  use_space_char: False
+  save_res_path: ./output/rec/predicts_vitstr.txt
+
+
+Optimizer:
+  name: Adadelta
+  epsilon: 1.e-8
+  rho: 0.95
+  clip_norm: 5.0
+  lr:
+    learning_rate: 1.0
+
+Architecture:
+  model_type: rec
+  algorithm: ViTSTR
+  in_channels: 1
+  Transform:
+  Backbone:
+    name: ViTSTR
+    scale: tiny
+  Neck:
+    name: SequenceEncoder
+    encoder_type: reshape
+  Head:
+    name: CTCHead
+
+Loss:
+  name: CELoss
+  with_all: True
+  ignore_index: &ignore_index 0 # Must be zero or greater than the number of character classes
+
+PostProcess:
+  name: ViTSTRLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/training/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ViTSTRLabelEncode: # Class handling label
+          ignore_index: *ignore_index
+      - GrayRecResizeImg:
+          image_shape: [224, 224] # W H
+          resize_type: PIL # PIL or OpenCV
+          inter_type: 'Image.BICUBIC'
+          scale: false
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 48
+    drop_last: True
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/evaluation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ViTSTRLabelEncode: # Class handling label
+          ignore_index: *ignore_index
+      - GrayRecResizeImg:
+          image_shape: [224, 224] # W H
+          resize_type: PIL # PIL or OpenCV
+          inter_type: 'Image.BICUBIC'
+          scale: false
+      - 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: 2