使用识别模型训练输入python tools/train.py -c pretrain_models/ch_PP-OCRv3_rec_distillation.yml，报错IndexError: list index out of range，请帮忙看下哪里出现问题呢

[bobo970722]

🔎 Search before asking

• I have searched the PaddleOCR Docs and found no similar bug report.
• I have searched the PaddleOCR Issues and found no similar bug report.
• I have searched the PaddleOCR Discussions and found no similar bug report.

🐛 Bug (问题描述)

部分训练集如下：
train_data/rec/train/000012.jpg 显卡:NVIDIA® GeForce RTX™ 4060
train_data/rec/train/000013.jpg 内存容量:32GB
train_data/rec/train/000014.jpg 处理器:Intel® Core™ Ultra 5 processor 125H
train_data/rec/train/000015.jpg 显卡:Intel® Arc™ graphics
train_data/rec/train/000016.jpg 显卡:Intel®Iris®Xe Graphics
train_data/rec/train/000017.jpg 屏幕:16英寸 FHD+
train_data/rec/train/000018.jpg 6 941812 767665

报错信息如下：
Message: 'When parsing line train_data/rec/train/000279.jpg Intel® Graphics\r\n, error happened with msg: Traceback (most recent call last):\n File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem\n label = substr[1]\nIndexError: list index out of range\n'
Arguments: ()
[2025/03/23 08:37:23] ppocr ERROR: When parsing line train_data/rec/train/000096.jpg 显卡:Intel® Graphics
, error happened with msg: Traceback (most recent call last):
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem
label = substr[1]
IndexError: list index out of range

--- Logging error ---
Traceback (most recent call last):
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem
label = substr[1]
IndexError: list index out of range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "D:\python3.10.0\lib\logging_init_.py", line 1101, in emit
stream.write(msg + self.terminator)
UnicodeEncodeError: 'gbk' codec can't encode character '\xae' in position 95: illegal multibyte sequence
Call stack:
File "D:\python3.10.0\lib\threading.py", line 966, in _bootstrap
self._bootstrap_inner()
File "D:\python3.10.0\lib\threading.py", line 1009, in _bootstrap_inner
self.run()
File "D:\python3.10.0\lib\threading.py", line 946, in run
self._target(*self._args, **self._kwargs)
File "D:\python3.10.0\lib\site-packages\paddle\fluid\dataloader\dataloader_iter.py", line 217, in _thread_loop
batch = self._dataset_fetcher.fetch(indices,
File "D:\python3.10.0\lib\site-packages\paddle\fluid\dataloader\fetcher.py", line 125, in fetch
data.append(self.dataset[idx])
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 169, in getitem
return self.getitem(rnd_idx)
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 169, in getitem
return self.getitem(rnd_idx)
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 169, in getitem
return self.getitem(rnd_idx)
[Previous line repeated 146 more times]
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 161, in getitem
self.logger.error(
Message: 'When parsing line train_data/rec/train/000096.jpg 显卡:Intel® Graphics\r\n, error happened with msg: Traceback (most recent call last):\n File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem\n label = substr[1]\nIndexError: list index out of range\n'
Arguments: ()
[2025/03/23 08:37:23] ppocr ERROR: When parsing line train_data/rec/train/000125.jpg 颜色:
, error happened with msg: Traceback (most recent call last):
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem
label = substr[1]
IndexError: list index out of range

[2025/03/23 08:37:23] ppocr ERROR: When parsing line train_data/rec/train/000032.jpg 处理器:Intel® Core™ i5-13420H Processor
, error happened with msg: Traceback (most recent call last):
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem
label = substr[1]
IndexError: list index out of range

--- Logging error ---
Traceback (most recent call last):
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem
label = substr[1]
IndexError: list index out of range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "D:\python3.10.0\lib\logging_init_.py", line 1101, in emit
stream.write(msg + self.terminator)
UnicodeEncodeError: 'gbk' codec can't encode character '\xae' in position 96: illegal multibyte sequence
Call stack:
File "D:\python3.10.0\lib\threading.py", line 966, in _bootstrap
self._bootstrap_inner()
File "D:\python3.10.0\lib\threading.py", line 1009, in _bootstrap_inner
self.run()
File "D:\python3.10.0\lib\threading.py", line 946, in run
self._target(*self._args, **self._kwargs)
File "D:\python3.10.0\lib\site-packages\paddle\fluid\dataloader\dataloader_iter.py", line 217, in _thread_loop
batch = self._dataset_fetcher.fetch(indices,
File "D:\python3.10.0\lib\site-packages\paddle\fluid\dataloader\fetcher.py", line 125, in fetch
data.append(self.dataset[idx])
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 169, in getitem
return self.getitem(rnd_idx)
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 169, in getitem
return self.getitem(rnd_idx)
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 169, in getitem
return self.getitem(rnd_idx)
[Previous line repeated 148 more times]
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 161, in getitem
self.logger.error(
Message: 'When parsing line train_data/rec/train/000032.jpg 处理器:Intel® Core™ i5-13420H Processor\r\n, error happened with msg: Traceback (most recent call last):\n File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem\n label = substr[1]\nIndexError: list index out of range\n'
Arguments: ()
[2025/03/23 08:37:23] ppocr ERROR: When parsing line train_data/rec/train/000282.jpg 净重:
, error happened with msg: Traceback (most recent call last):
File "D:\PaddleOCR-release-2.7\ppocr\data\simple_dataset.py", line 150, in getitem
label = substr[1]
IndexError: list index out of range

🏃‍♂️ Environment (运行环境)

os windows 11
environment
python 3.10.0
paddleocr 2.7
install zip
ram 32.0 GB
cpu Intel(R) Core(TM) Ultra 7 155H

🌰 Minimal Reproducible Example (最小可复现问题的Demo)

部分训练集如下：
train_data/rec/train/000012.jpg 显卡:NVIDIA® GeForce RTX™ 4060
train_data/rec/train/000013.jpg 内存容量:32GB
train_data/rec/train/000014.jpg 处理器:Intel® Core™ Ultra 5 processor 125H
train_data/rec/train/000015.jpg 显卡:Intel® Arc™ graphics
train_data/rec/train/000016.jpg 显卡:Intel®Iris®Xe Graphics
train_data/rec/train/000017.jpg 屏幕:16英寸 FHD+
train_data/rec/train/000018.jpg 6 941812 767665

[cuicheng01]

6 941812 767665中的空格是“\t”还是空格呢？

[bobo970722]

6 941812 767665中的空格是“\t”还是空格呢？
是空格，我在txt文件中检索不到\t存在

[GreatV]

你在运行 Python tools/train.py -c pretrain_models/ch_PP-OCRv3_rec_distillation.yml 进行文本识别模型训练时报错：

IndexError: list index out of range，该错误多次出现在 simple_dataset.py 第 150 行 label = substr[1]，表示数据解析失败，主要原因在于你的训练数据格式不符合预期。

问题原因分析：

PaddleOCR 识别模型的数据格式通常是 txt 文件，每行一条数据，格式如下：

图像路径标签内容

你提供的训练数据例如：

train_data/rec/train/000012.jpg 显卡:NVIDIA® GeForce RTX™ 4060

看似格式正确，实则问题可能有：

1. 部分样本的标签为空：
   例如下面这行：
   train_data/rec/train/000125.jpg 颜色:

   标签“颜色:”后面为空，split 使用空格或tab 拆分后 substr[1] 不存在，造成 IndexError。

2. 使用了非标准 TAB 分隔符：
   如果文件中使用的是多个空格而不是制表符（TAB \t ），可能导致 split 分割失败。

3. 部分标签含有特殊字符或换行：
   比如 Intel®、™、\r\n 等符号会妨碍日志打印（报错 UnicodeEncodeError: 'gbk' codec can't encode character ），也有可能影响读取。

建议解决方案：

1. 检查并清洗你的训练标注文件（例如 train.txt 或类似的数据列表）：

• 确保每一行都是严格的「图片路径 + 制表符 + 标签内容」（必须用制表符\t，而不是空格）。

• 标签不能为空。

• 标签内容中避免换行或不可见字符。

• 对文件做如下清洗示例：

  import re

  with open("train_label.txt", "r", encoding="utf-8") as fin:
  with open("train_clean.txt", "w", encoding="utf-8") as fout:
  for line in fin:
  line = line.strip()
  if not line:
  continue
  parts = re.split(r"\s+", line, maxsplit=1)
  if len(parts) != 2 or not parts[1].strip():
  continue # 跳过格式不对或标签为空的行
  fout.write(parts[0] + "\t" + parts[1].strip() + "\n")

2. 修改配置 yml 文件中数据路径：

确保它引用的是你清洗过的新标签文件：

例如：

Train:
dataset:
name: SimpleDataSet
data_dir: ./train_data/rec
label_file_list: ["./train_data/rec/train_clean.txt"]
transforms:
...

3. 若数据源是图像+标签字典，建议转成标准格式。

参考 Issue：

此问题与 PaddleOCR 官方 issue #13826 有关：#13826

该 issue 同样反映了 label 格式异常导致 IndexError 的情况。

4. 编码问题：

如果你在 Windows 上运行，控制台默认编码是 gbk，而你的标签文本中含有 Unicode 字符（如®），建议：

训练时加入环境变量，避免编码错误影响日志：

set PYTHONIOENCODING=utf-8

或将 logging 配置为输出 UTF-8。

总结：

你遇到的 IndexError: list index out of range 错误，很可能是由于训练列表文件中存在格式异常行，例如标签为空、使用空格分隔、或其他非法字符。请重点清查你的标注文件并按上述格式清洗后重新训练。

---

Response generated by 🤖 feifei-bot | chatgpt-4o-latest

⚠️ This is an AI-generated response. Please verify any critical information.