PaddlePaddle/PaddleOCR · error · ValueError

{model_name} is not supported. Please check if the model is

Error message

{model_name} is not supported. Please check if the model is supported by the PaddleOCR wheel.

What it means

TextClassifier (tools/infer/predict_cls.py) reads inference.yml from the cls model directory and, when its Global.model_name is set but not in the wheel-supported whitelist (PP-LCNet_x1_0_textline_ori, PP-LCNet_x0_25_textline_ori), raises ValueError. This is a compatibility gate: the pip-installed PaddleOCR wheel only ships support for specific textline-orientation classifiers, and self-trained or third-party models are rejected up front. If inference.yml is absent, the check is skipped entirely.

Source

Thrown at tools/infer/predict_cls.py:47

import tools.infer.utility as utility
from ppocr.postprocess import build_post_process
from ppocr.utils.logging import get_logger
from ppocr.utils.utility import get_image_file_list, check_and_read

logger = get_logger()


class TextClassifier(object):
    def __init__(self, args):
        if os.path.exists(f"{args.cls_model_dir}/inference.yml"):
            model_config = utility.load_config(f"{args.cls_model_dir}/inference.yml")
            model_name = model_config.get("Global", {}).get("model_name", "")
            if model_name and model_name not in [
                "PP-LCNet_x1_0_textline_ori",
                "PP-LCNet_x0_25_textline_ori",
            ]:
                raise ValueError(
                    f"{model_name} is not supported. Please check if the model is supported by the PaddleOCR wheel."
                )

        self.cls_image_shape = [int(v) for v in args.cls_image_shape.split(",")]
        self.cls_batch_num = args.cls_batch_num
        self.cls_thresh = args.cls_thresh
        postprocess_params = {
            "name": "ClsPostProcess",
            "label_list": args.label_list,
        }
        self.postprocess_op = build_post_process(postprocess_params)
        (
            self.predictor,
            self.input_tensor,
            self.output_tensors,
            _,
        ) = utility.create_predictor(args, "cls", logger)
        self.use_onnx = args.use_onnx

View on GitHub (pinned to 2661c7c0ef)

Solutions

  1. Download the supported PP-LCNet_x1_0_textline_ori / PP-LCNet_x0_25_textline_ori cls model from the official PaddleOCR model repo
  2. Edit inference.yml in the custom model dir and set Global.model_name to a whitelisted name (only if the model is architecturally identical)
  3. If the model has no inference.yml requirement, ensure the dir actually contains one — or remove inference.yml so the gate is skipped (custom models at your own risk)

Example fix

# before
python -m tools.infer.utility --cls_model_dir=./my_cls
# ValueError: cls_mobile_v1.0 is not supported...

# after
wget https://paddleocr.bj.bcebos.com/.../PP-LCNet_x0_25_textline_ori_train.tar
tar -xf PP-LCNet_x0_25_textline_ori_train.tar -C inference/
python tools/infer/predict_system.py --image_dir=... --cls_model_dir=inference/PP-LCNet_x0_25_textline_ori
Defensive patterns

Strategy: validation

Validate before calling

import os, yaml
SUPPORTED_CLS = {'PP-LCNet_x1_0_textline_ori', 'PP-LCNet_x0_25_textline_ori'}
yml = os.path.join(cls_model_dir, 'inference.yml')
if os.path.exists(yml):
    name = yaml.safe_load(open(yml)).get('Global', {}).get('model_name', '')
    assert not name or name in SUPPORTED_CLS, f'cls model {name!r} unsupported by wheel'

Try / catch

try:
    cls_predictor = TextClassifier(args)
except ValueError as e:
    if 'not supported' in str(e):
        logger.error('switch to a wheel-supported textline-ori model')
        raise SystemExit(2)
    raise

Prevention

When it happens

Trigger: Pointing --cls_model_dir at a self-trained or downloaded cls model whose inference.yml declares a different model_name (e.g. 'cls_ch', 'PP-LCNet_x0_5_textline_ori', or a custom name).

Common situations: Users switching from the old GitHub-repo layout (models without inference.yml, which pass silently) to the pip wheel; downloading cls models from PaddleOCR model zoo pages that predate the whitelist; training a custom angle classifier and reusing it with the wheel.

Related errors

AI-assisted analysis of PaddlePaddle/PaddleOCR@2661c7c0ef (2026-08-14). Data as JSON: /api/errors/1564d6cc2a872d56. Report an issue: GitHub.