PaddleOCR V3 训练 ValueError: too many values to unpack (expected 4)

[youhong199]

[2025/06/26 15:14:51] ppocr INFO: During the training process, after the 0th iteration, an evaluation is run every 5 iterations
Traceback (most recent call last):
File "C:\Program Files\JetBrains\PyCharm 2025.1.2\plugins\python-ce\helpers\pydev\pydevd.py", line 1570, in _exec
pydev_imports.execfile(file, globals, locals) # execute the script
File "C:\Program Files\JetBrains\PyCharm 2025.1.2\plugins\python-ce\helpers\pydev_pydev_imps_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "tools/train.py", line 273, in
main(config, device, logger, vdl_writer, seed)
File "tools/train.py", line 225, in main
program.train(
File "C:\Users\Admin\Desktop\Tool\Paddle\PaddleOCR\tools\program.py", line 421, in train
post_result = post_process_class(preds, batch[1])
File "C:\Users\Admin\Desktop\Tool\Paddle\PaddleOCR\ppocr\postprocess\db_postprocess.py", line 288, in call
results[k] = self.post_process(predicts[k], shape_list=shape_list)
File "C:\Users\Admin\Desktop\Tool\Paddle\PaddleOCR\ppocr\postprocess\db_postprocess.py", line 236, in call
src_h, src_w, ratio_h, ratio_w = shape_list[batch_index]
ValueError: too many values to unpack (expected 4)
python-BaseException
Traceback (most recent call last):
File "C:\Users\Admin\AppData\Local\Programs\Python\Python39\lib\subprocess.py", line 524, in run
raise CalledProcessError(retcode, process.args,

PP-OCRv3_det_cml.yml
Eval:
dataset:
name: SimpleDataSet
data_dir: ./train_data/val_image/val_img/
label_file_list:
- ./train_data/val_image/val_label/val_label.txt
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- DetLabelEncode: # Class handling label
- DetResizeForTest:
image_shape: [ 960, 960 ]
resize_type: 0
- NormalizeImage:
scale: 1./255.
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: 'hwc'
- ToCHWImage:
- KeepKeys:
keep_keys: ['image', 'shape', 'polys', 'ignore_tags']

[youhong199]

❗ Error when using custom evaluation and training KeepKeys settings
When I modified the KeepKeys settings in both the Eval and Train parts of my YAML config file to include 'shape', I encountered a runtime error.

Here is the log output:

lua
复制
编辑
[2025/06/26 15:47:18] ppocr ERROR: When parsing line 0019.jpg [{"transcription": "吕课", "points": [[95, 57], [137, 57], [137, 79], [95, 79]], "difficult": false, "key_cls": "name"}, {"transcription": "性别未知民族汉", "points": [[46, 87], [210, 87], [210, 110], [46, 110]], "difficult": false, "key_cls": "sex"}, {"transcription": "2014年1月16日", "points": [[91, 116], [240, 116], [240, 140], [91, 140]], "difficult": false, "key_cls": "birth"}, {"transcription": "山东省济南某某666号—自制数据集", "points": [[92, 151], [280, 151], [280, 194], [92, 194]], "difficult": false, "key_cls": "adress"}, {"transcription": "418547080384305138", "points": [[146, 235], [310, 235], [310, 260], [146, 260]], "difficult": false, "key_cls": "card"}]
, error happened with msg: Traceback (most recent call last):
File "C:\Users\Admin\Desktop\Tool\Paddle\PaddleOCR\ppocr\data\simple_dataset.py", line 137, in getitem
outs = transform(data, self.ops)
File "C:\Users\Admin\Desktop\Tool\Paddle\PaddleOCR\ppocr\data\imaug_init_.py", line 73, in transform
data = op(data)
File "C:\Users\Admin\Desktop\Tool\Paddle\PaddleOCR\ppocr\data\imaug\operators.py", line 123, in call
data_list.append(data[key])
KeyError: 'shape'
Additionally, I encountered:

pgsql
复制
编辑
IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed
I suspect this may have been caused by a change I made based on a suggestion from rudra0713 on May 22, 2024, where I modified KeepKeys in the YAML file like below:

yaml

• KeepKeys:
  keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
  This seems to break the pipeline at runtime.

Any help or suggestions on how to fix this and properly include shape in KeepKeys during evaluation and training would be appreciated.

Global:
debug: true
use_gpu: false
epoch_num: 20
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/ch_PP-OCR_v3_det/
save_epoch_step: 100
eval_batch_step:

• 0
• 5
  cal_metric_during_train: true
  pretrained_model: null
  checkpoints: null
  save_inference_dir: null
  use_visualdl: false
  pretrain_weights: ./ch_PP-OCRv3_det_slim_distill_train/best_accuracy.pdparams
  infer_img: doc/imgs_en/img_10.jpg
  save_res_path: ./checkpoints/det_db/predicts_db.txt
  distributed: true
  d2s_train_image_shape: [3, -1, -1]
  amp_dtype: bfloat16

Architecture:
name: DistillationModel
algorithm: Distillation
model_type: det
Models:
Student:
pretrained: ./ch_PP-OCRv3_det_slim_distill_train/best_accuracy.pdparams
model_type: det
algorithm: DB
Transform: null
Backbone:
name: MobileNetV3
scale: 0.5
model_name: large
disable_se: true
Neck:
name: RSEFPN
out_channels: 96
shortcut: True
Head:
name: DBHead
k: 50
Student2:
pretrained: ./ch_PP-OCRv3_det_slim_distill_train/best_accuracy.pdparams
model_type: det
algorithm: DB
Transform: null
Backbone:
name: MobileNetV3
scale: 0.5
model_name: large
disable_se: true
Neck:
name: RSEFPN
out_channels: 96
shortcut: True
Head:
name: DBHead
k: 50
Teacher:
freeze_params: true
return_all_feats: false
model_type: det
algorithm: DB
Backbone:
name: ResNet_vd
in_channels: 3
layers: 50
Neck:
name: LKPAN
out_channels: 256
Head:
name: DBHead
kernel_list: [7,2,2]
k: 50

Loss:
name: CombinedLoss
loss_config_list:

• DistillationDilaDBLoss:
  weight: 1.0
  model_name_pairs:
  • ["Student", "Teacher"]
  • ["Student2", "Teacher"]
    key: maps
    balance_loss: true
    main_loss_type: DiceLoss
    alpha: 5
    beta: 10
    ohem_ratio: 3
• DistillationDMLLoss:
  model_name_pairs:
  • ["Student", "Student2"]
    maps_name: "thrink_maps"
    weight: 1.0
    model_name_pairs: ["Student", "Student2"]
    key: maps
• DistillationDBLoss:
  weight: 1.0
  model_name_list: ["Student", "Student2"]
  balance_loss: true
  main_loss_type: DiceLoss
  alpha: 5
  beta: 10
  ohem_ratio: 3

Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.001
warmup_epoch: 2
regularizer:
name: L2
factor: 5.0e-05

PostProcess:
name: DistillationDBPostProcess
model_name: ["Student","Student2"]
key: head_out
thresh: 0.3
box_thresh: 0.6
max_candidates: 1000
unclip_ratio: 1.5

Metric:
name: DistillationMetric
base_metric_name: DetMetric
main_indicator: hmean
key: "Student"

Train:
dataset:
name: SimpleDataSet
data_dir: ./train_data/train_image/train_img/
label_file_list:
- ./train_data/train_image/train_label/train_label.txt
ratio_list: [1.0]
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- DetLabelEncode: null
- CopyPaste:
- IaaAugment:
augmenter_args:
- type: Fliplr
args:
p: 0.5
- type: Affine
args:
rotate:
- -10
- 10
- type: Resize
args:
size:
- 0.5
- 3
- EastRandomCropData:
size:
- 960
- 960
max_tries: 50
keep_ratio: true
- MakeBorderMap:
shrink_ratio: 0.4
thresh_min: 0.3
thresh_max: 0.7
- MakeShrinkMap:
shrink_ratio: 0.4
min_text_size: 8
- NormalizeImage:
scale: 1./255.
mean:
- 0.485
- 0.456
- 0.406
std:
- 0.229
- 0.224
- 0.225
order: hwc
- ToCHWImage: null
- KeepKeys:
keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
loader:
shuffle: true
drop_last: false
batch_size_per_card: 2
num_workers: 1

Eval:
dataset:
name: SimpleDataSet
data_dir: ./train_data/val_image/val_img/
label_file_list:
- ./train_data/val_image/val_label/val_label.txt
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- DetLabelEncode: # Class handling label
- DetResizeForTest:
image_shape: [ 960, 960 ]
resize_type: 0
- NormalizeImage:
scale: 1./255.
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: 'hwc'
- ToCHWImage:
- KeepKeys:
keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
loader:
shuffle: False
drop_last: False
batch_size_per_card: 1 # must be 1
num_workers: 0