EasyData/deploy/configs/ppdataaug_ocr_text2img.yaml at develop · andyjiang1116/EasyData · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
DataGen:
  mode: text2img
  config: "deploy/configs/ppdataaug/ocr_rec_gen.yaml"
  gen_num: 5
  out_dir: "test"
  bg_num_per_word: 5
  threads: 1
  bg_img_dir: "demo/ocr_rec/bg/"
  font_dir: "demo/ocr_rec/font"
  corpus_file: "demo/ocr_rec/corpus.txt"
  img_save_folder: &img_save_folder output_img
  gen_label: &gen_label output_img/all_label.txt
  delimiter: &delimiter "\t"

FeatureExtract:
  config: "deploy/configs/ppcv/feature_extract.yaml"
  thresh: 0.9
  file_out: "tmp/rm_repeat.txt"

  # indexing engine config
IndexProcess:
  index_method: "HNSW32" # supported: HNSW32, IVF, Flat
  image_root: *img_save_folder
  index_dir: "./augdata/all_aug"
  all_label_file:  *gen_label
  index_operation: "new" # suported: "append", "remove", "new"
  delimiter: *delimiter
  dist_type: "IP"
  embedding_size: 512
  batch_size: 32
  return_k: 5
  score_thres: 0.5

BigModel:
  model_type: ocr_rec  # support(cls / ocr_rec)
  config: "deploy/configs/ppcv/text_recognition.yaml"
  batch_size: 8
  thresh: 0.1
  final_label: "high_socre_label.txt"