openocr-python 0.0.2__py3-none-any.whl

openocr/configs/rec/gtc/svtrv2_lnconv_smtr_gtc_long_infer.yml

@@ -0,0 +1,151 @@
+Global:
+  device: gpu
+  epoch_num: 20
+  log_smooth_window: 20
+  print_batch_step: 10
+  output_dir: ./output/rec/svtrv2_lnconv_smtr_gtc_long_infer
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 1000]
+  eval_epoch_step: [0, 1]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  use_tensorboard: false
+  infer_img: ../ltb/img
+  # for data or label process
+  character_dict_path: &character_dict_path ./tools/utils/EN_symbol_dict.txt # 96en
+  # ./tools/utils/ppocr_keys_v1.txt  # ch
+  max_text_length: &max_text_length 25
+  use_space_char: &use_space_char False
+  save_res_path: ./output/rec/predicts_smtr.txt
+  use_amp: True
+  distributed: true
+
+Optimizer:
+  name: AdamW
+  lr: 0.000325
+  weight_decay: 0.05
+  filter_bias_and_bn: True
+
+LRScheduler:
+  name: OneCycleLR
+  warmup_epoch: 1.5 # pct_start 0.075*20 = 1.5ep
+  cycle_momentum: False
+
+Architecture:
+  model_type: rec
+  algorithm: BGPD
+  in_channels: 3
+  Transform:
+  Encoder:
+    name: SVTRv2LNConvTwo33
+    use_pos_embed: False
+    out_channels: 256
+    dims: [128, 256, 384]
+    depths: [6, 6, 6]
+    num_heads: [4, 8, 12]
+    mixer: [['Conv','Conv','Conv','Conv','Conv','Conv'],['Conv','Conv','FGlobal','Global','Global','Global'],['Global','Global','Global','Global','Global','Global']]
+    local_k: [[5, 5], [5, 5], [-1, -1]]
+    sub_k: [[1, 1], [2, 1], [-1, -1]]
+    last_stage: false
+    feat2d: True
+  Decoder:
+    name: GTCDecoder
+    infer_gtc: False
+    detach: False
+    gtc_decoder:
+      name: SMTRDecoder
+      num_layer: 1
+      ds: True
+      max_len: *max_text_length
+      next_mode: &next True
+      sub_str_len: &subsl 5
+    ctc_decoder:
+      name: RCTCDecoder
+
+Loss:
+  name: CTCLoss
+
+PostProcess:
+  name: CTCLabelDecode
+  character_dict_path: *character_dict_path
+  use_space_char: *use_space_char
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+  is_filter: True
+
+Train:
+  dataset:
+    name: RatioDataSetTVResize
+    ds_width: True
+    padding: false
+    data_dir_list: ['../Union14M-L-LMDB-Filtered/filter_train_challenging',
+    '../Union14M-L-LMDB-Filtered/filter_train_hard',
+    '../Union14M-L-LMDB-Filtered/filter_train_medium',
+    '../Union14M-L-LMDB-Filtered/filter_train_normal',
+    '../Union14M-L-LMDB-Filtered/filter_train_easy',
+    ]
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - PARSeqAugPIL:
+      - CTCLabelEncode: # Class handling label
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: &bs 128
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: True
+  loader:
+    shuffle: True
+    batch_size_per_card: *bs
+    drop_last: True
+    max_ratio: &max_ratio 12
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: RatioDataSetTVResize
+    ds_width: True
+    padding: False
+    data_dir_list: [
+      '../evaluation/CUTE80',
+      '../evaluation/IC13_857',
+      '../evaluation/IC15_1811',
+      '../evaluation/IIIT5k',
+      '../evaluation/SVT',
+      '../evaluation/SVTP',
+      ]
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - CTCLabelEncode: # Class handling label
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: *bs
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: False
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: *bs
+    max_ratio: *max_ratio
+    num_workers: 4

openocr/configs/rec/gtc/svtrv2_lnconv_smtr_gtc_smtr_long.yml

@@ -0,0 +1,150 @@
+Global:
+  device: gpu
+  epoch_num: 20
+  log_smooth_window: 20
+  print_batch_step: 10
+  output_dir: ./output/rec/svtrv2_lnconv_smtr_gtc_nodetach_smtr_long_infer
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 1000]
+  eval_epoch_step: [0, 1]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  use_tensorboard: false
+  infer_img:
+  # for data or label process
+  character_dict_path: &character_dict_path ./tools/utils/EN_symbol_dict.txt # 96en
+  # ./tools/utils/ppocr_keys_v1.txt  # ch
+  max_text_length: &max_text_length 25
+  use_space_char: &use_space_char False
+  save_res_path: ./output/rec/predicts_smtr.txt
+  use_amp: True
+  distributed: true
+
+Optimizer:
+  name: AdamW
+  lr: 0.000325
+  weight_decay: 0.05
+  filter_bias_and_bn: True
+
+LRScheduler:
+  name: OneCycleLR
+  warmup_epoch: 1.5 # pct_start 0.075*20 = 1.5ep
+  cycle_momentum: False
+Architecture:
+  model_type: rec
+  algorithm: BGPD
+  in_channels: 3
+  Transform:
+  Encoder:
+    name: SVTRv2LNConvTwo33
+    use_pos_embed: False
+    out_channels: 256
+    dims: [128, 256, 384]
+    depths: [6, 6, 6]
+    num_heads: [4, 8, 12]
+    mixer: [['Conv','Conv','Conv','Conv','Conv','Conv'],['Conv','Conv','FGlobal','Global','Global','Global'],['Global','Global','Global','Global','Global','Global']]
+    local_k: [[5, 5], [5, 5], [-1, -1]]
+    sub_k: [[1, 1], [2, 1], [-1, -1]]
+    last_stage: false
+    feat2d: True
+  Decoder:
+    name: GTCDecoder
+    infer_gtc: True
+    detach: False
+    gtc_decoder:
+      name: SMTRDecoder
+      num_layer: 1
+      ds: True
+      max_len: *max_text_length
+      next_mode: &next True
+      sub_str_len: &subsl 5
+      infer_aug: True
+    ctc_decoder:
+      name: RCTCDecoder
+
+Loss:
+  name: GTCLoss
+  ctc_weight: 0.1
+  gtc_loss:
+    name: SMTRLoss
+
+PostProcess:
+  name: GTCLabelDecode
+  gtc_label_decode:
+    name: SMTRLabelDecode
+    next_mode: *next
+  character_dict_path: *character_dict_path
+  use_space_char: *use_space_char
+  only_gtc: True
+
+Metric:
+  name: RecGTCMetric
+  main_indicator: acc
+  is_filter: True
+
+Train:
+  dataset:
+    name: RatioDataSetTVResize
+    ds_width: True
+    padding: false
+    data_dir_list: ['../Union14M-L-LMDB-Filtered/filter_train_challenging',
+    '../Union14M-L-LMDB-Filtered/filter_train_hard',
+    '../Union14M-L-LMDB-Filtered/filter_train_medium',
+    '../Union14M-L-LMDB-Filtered/filter_train_normal',
+    '../Union14M-L-LMDB-Filtered/filter_train_easy',
+    ]
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - PARSeqAugPIL:
+      - SMTRLabelEncode: # Class handling label
+          sub_str_len: *subsl
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'label_subs', 'label_next', 'length_subs',
+          'label_subs_pre', 'label_next_pre', 'length_subs_pre', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: &bs 256
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: True
+  loader:
+    shuffle: True
+    batch_size_per_card: *bs
+    drop_last: True
+    max_ratio: &max_ratio 12
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ../ltb/
+    label_file_list: ['../ltb/ultra_long_70_list.txt']
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - GTCLabelEncode: # Class handling label
+          gtc_label_encode:
+            name: ARLabelEncode
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: 200
+      - SliceResize:
+          image_shape: [3, 32, 128]
+          padding: False
+          max_ratio: 12
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length', 'ctc_label', 'ctc_length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1
+    num_workers: 2

openocr/configs/rec/gtc/svtrv2_lnconv_smtr_gtc_stream.yml

@@ -0,0 +1,152 @@
+Global:
+  device: gpu
+  epoch_num: 60
+  log_smooth_window: 20
+  print_batch_step: 10
+  output_dir: ./output/rec/svtrv2_lnconv_smtr_gtc_stream
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 500]
+  eval_epoch_step: [0, 1]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  use_tensorboard: false
+  infer_img:
+  # for data or label process
+  character_dict_path: &character_dict_path ./tools/utils/EN_symbol_dict.txt # 96en
+  # ./tools/utils/ppocr_keys_v1.txt  # ch
+  max_text_length: &max_text_length 25
+  use_space_char: &use_space_char False
+  save_res_path: ./output/rec/predicts_smtr.txt
+  use_amp: True
+  distributed: true
+  grad_clip_val: 20
+
+Optimizer:
+  name: AdamW
+  lr: 0.00065
+  weight_decay: 0.05
+  filter_bias_and_bn: True
+
+LRScheduler:
+  name: OneCycleLR
+  warmup_epoch: 5 # pct_start 0.075*20 = 1.5ep
+  cycle_momentum: False
+
+Architecture:
+  model_type: rec
+  algorithm: BGPD
+  in_channels: 3
+  Transform:
+  Encoder:
+    name: SVTRv2LNConvTwo33
+    use_pos_embed: False
+    out_channels: 256
+    dims: [128, 256, 384]
+    depths: [6, 6, 6]
+    num_heads: [4, 8, 12]
+    mixer: [['Conv','Conv','Conv','Conv','Conv','Conv'],['Conv','Conv','FGlobal','Global','Global','Global'],['Global','Global','Global','Global','Global','Global']]
+    local_k: [[5, 5], [5, 5], [-1, -1]]
+    sub_k: [[1, 1], [2, 1], [-1, -1]]
+    last_stage: false
+    feat2d: True
+  Decoder:
+    name: GTCDecoder
+    infer_gtc: True
+    detach: False
+    gtc_decoder:
+      name: SMTRDecoder
+      num_layer: 1
+      ds: True
+      max_len: *max_text_length
+      next_mode: &next True
+      sub_str_len: &subsl 5
+      infer_aug: False
+    ctc_decoder:
+      name: RCTCDecoder
+
+Loss:
+  name: GTCLoss
+  ctc_weight: 0.25
+  gtc_loss:
+    name: SMTRLoss
+
+PostProcess:
+  name: GTCLabelDecode
+  gtc_label_decode:
+    name: SMTRLabelDecode
+    next_mode: *next
+  character_dict_path: *character_dict_path
+  use_space_char: *use_space_char
+  only_gtc: True
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+  is_filter: True
+  stream: True
+
+Train:
+  dataset:
+    name: RatioDataSetTVResize
+    ds_width: True
+    padding: false
+    data_dir_list: ['../Union14M-L-LMDB-Filtered/filter_train_challenging',
+    '../Union14M-L-LMDB-Filtered/filter_train_hard',
+    '../Union14M-L-LMDB-Filtered/filter_train_medium',
+    '../Union14M-L-LMDB-Filtered/filter_train_normal',
+    '../Union14M-L-LMDB-Filtered/filter_train_easy',
+    ]
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - PARSeqAugPIL:
+      - SMTRLabelEncode: # Class handling label
+          sub_str_len: *subsl
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'label_subs', 'label_next', 'length_subs',
+          'label_subs_pre', 'label_next_pre', 'length_subs_pre', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: &bs 256
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: True
+  loader:
+    shuffle: True
+    batch_size_per_card: *bs
+    drop_last: True
+    max_ratio: &max_ratio 12
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ../ltb/
+    label_file_list: ['../ltb/ultra_long_70_list.txt']
+    transforms:
+      - DecodeImagePIL: # load image
+          img_mode: RGB
+      - GTCLabelEncode: # Class handling label
+          gtc_label_encode:
+            name: ARLabelEncode
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - SliceTVResize:
+          image_shape: [32, 128]
+          padding: False
+          max_ratio: 4
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length', 'ctc_label', 'ctc_length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1
+    num_workers: 2

openocr/configs/rec/igtr/svtr_base_ds_igtr.yml

@@ -0,0 +1,157 @@
+Global:
+  device: gpu
+  epoch_num: 20
+  log_smooth_window: 20
+  print_batch_step: 10
+  output_dir: ./output/rec/u14m_filter/svtr_base_igtr
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 500]
+  eval_epoch_step: [0, 1]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  use_tensorboard: false
+  infer_img:
+  # for data or label process
+  character_dict_path: &character_dict_path
+  # ./tools/utils/EN_symbol_dict.txt # 96en
+  # ./tools/utils/ppocr_keys_v1.txt  # ch
+  max_text_length: &max_text_length 25
+  use_space_char: &use_space_char False
+  save_res_path: ./output/rec/u14m_filter/predicts_svtr_base_igtr.txt
+  use_amp: True
+
+Optimizer:
+  name: AdamW
+  lr: 0.0005 # 2gpus 384bs/gpu
+  weight_decay: 0.05
+  filter_bias_and_bn: True
+
+LRScheduler:
+  name: OneCycleLR
+  warmup_epoch: 1.5
+  cycle_momentum: False
+
+Architecture:
+  model_type: rec
+  algorithm: IGTR
+  in_channels: 3
+  Transform:
+  Encoder:
+    name: SVTRNet2DPos
+    img_size: [32, -1]
+    out_char_num: 25
+    out_channels: 256
+    patch_merging: 'Conv'
+    embed_dim: [128, 256, 384]
+    depth: [6, 6, 6]
+    num_heads: [4, 8, 12]
+    mixer: ['ConvB','ConvB','ConvB','ConvB','ConvB','ConvB', 'ConvB','ConvB', 'Global','Global','Global','Global','Global','Global','Global','Global','Global','Global']
+    local_mixer: [[5, 5], [5, 5], [5, 5]]
+    last_stage: False
+    prenorm: True
+    use_first_sub: False
+  Decoder:
+    name: IGTRDecoder
+    dim: 384
+    num_layer: 1
+    ar: False
+    refine_iter: 0
+    # next_pred: True
+    next_pred: False
+    pos2d: True
+    ds: True
+    # pos_len: False
+    # rec_layer: 1
+
+
+Loss:
+  name: IGTRLoss
+
+PostProcess:
+  name: IGTRLabelDecode
+  character_dict_path: *character_dict_path
+  use_space_char: *use_space_char
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: RatioDataSet
+    ds_width: True
+    padding: &padding False
+    data_dir_list: ['../Union14M-L-LMDB-Filtered/filter_train_challenging',
+    '../Union14M-L-LMDB-Filtered/filter_train_hard',
+    '../Union14M-L-LMDB-Filtered/filter_train_medium',
+    '../Union14M-L-LMDB-Filtered/filter_train_normal',
+    '../Union14M-L-LMDB-Filtered/filter_train_easy',
+    ]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - PARSeqAug:
+      - IGTRLabelEncode: # Class handling label
+          k: 8
+          prompt_error: False
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'prompt_pos_idx_list',
+          'prompt_char_idx_list', 'ques_pos_idx_list', 'ques1_answer_list',
+          'ques2_char_idx_list', 'ques2_answer_list', 'ques3_answer', 'ques4_char_num_list',
+          'ques_len_list', 'ques2_len_list', 'prompt_len_list', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: &bs 384
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: True
+  loader:
+    shuffle: True
+    batch_size_per_card: *bs
+    drop_last: True
+    max_ratio: &max_ratio 4
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: RatioDataSet
+    ds_width: True
+    padding: *padding
+    data_dir_list: ['../evaluation/CUTE80',
+    '../evaluation/IC13_857',
+    '../evaluation/IC15_1811',
+    '../evaluation/IIIT5k',
+    '../evaluation/SVT',
+    '../evaluation/SVTP']
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ARLabelEncode: # Class handling label
+          character_dict_path: *character_dict_path
+          use_space_char: *use_space_char
+          max_text_length: *max_text_length
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  sampler:
+    name: RatioSampler
+    scales: [[128, 32]] # w, h
+    # divide_factor: to ensure the width and height dimensions can be devided by downsampling multiple
+    first_bs: 256
+    fix_bs: false
+    divided_factor: [4, 16] # w, h
+    is_training: False
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 256
+    max_ratio: *max_ratio
+    num_workers: 4