PaddlePaddle/PaddleOCR · error · TypeError

The input data is inconsistent with expectations.

Error message

The input data is inconsistent with expectations.

What it means

Raised as TypeError by the kie_ser hub module's predict() when the arguments match neither accepted shape: a non-empty images list with paths == [], or images == [] with a non-empty paths list. Passing both, neither, non-list values, or None (None != [] is True but isinstance(None, list) is False) all fail the exclusive-or style check. It is an input-contract error on the hub HTTP predict handler.

Source

Thrown at deploy/hubserving/kie_ser/module.py:115

                continue
            images.append(img)
        return images

    def predict(self, images=[], paths=[]):
        """
        Get the chinese texts in the predicted images.
        Args:
            images (list(numpy.ndarray)): images data, shape of each is [H, W, C]. If images not paths
            paths (list[str]): The paths of images. If paths not images
        Returns:
            res (list): The result of chinese texts and save path of images.
        """
        if images != [] and isinstance(images, list) and paths == []:
            predicted_data = images
        elif images == [] and isinstance(paths, list) and paths != []:
            predicted_data = self.read_images(paths)
        else:
            raise TypeError("The input data is inconsistent with expectations.")

        assert (
            predicted_data != []
        ), "There is not any image to be predicted. Please check the input data."

        all_results = []
        for img in predicted_data:
            if img is None:
                logger.info("error in loading image")
                all_results.append([])
                continue
            starttime = time.time()
            ser_res, _, elapse = self.ser_predictor(img)
            elapse = time.time() - starttime
            logger.info("Predict time: {}".format(elapse))
            all_results.append(ser_res)
        return all_results

View on GitHub (pinned to 2661c7c0ef)

Solutions

1. Send exactly one input form: {"images": [ndarray, ...], "paths": []} or {"images": [], "paths": ["/data/a.jpg", ...]}
2. Ensure both fields are JSON arrays (never null) in the request body
3. Decode base64 images to numpy arrays (cv2.imdecode) before calling the python API
4. Verify at least one element is present in the chosen list

Example fix

# before
res = module.predict(images=None, paths=None)  # TypeError

# after
res = module.predict(images=[], paths=["/data/invoice_01.jpg"])
# or
res = module.predict(images=[img_ndarray], paths=[])

def valid_predict_input(images, paths) -> bool:
    images_ok = isinstance(images, list) and len(images) > 0
    paths_ok = isinstance(paths, list) and len(paths) > 0
    return (images_ok and paths == []) or (paths_ok and images == [])

assert valid_predict_input(images, paths), "pass exactly one of images/paths as non-empty lists"

from typing import Any

def is_predict_payload(data: Any) -> bool:
    images = data.get("images") if isinstance(data, dict) else None
    paths = data.get("paths") if isinstance(data, dict) else None
    if not (isinstance(images, list) and isinstance(paths, list)):
        return False
    return bool(images) != bool(paths)  # exactly one non-empty

try:
    results = module.predict(images=images, paths=paths)
except TypeError as e:
    if "inconsistent with expectations" in str(e):
        # normalize inputs and retry with exactly one form
        results = module.predict(images=[], paths=[str(p) for p in paths or []])
    else:
        raise

When it happens

Trigger: POSTing {"images": null, "paths": null} or {"images": [...], "paths": [...]} to the hub endpoint; sending a single numpy array instead of a list; passing a string path in images; empty lists on both sides; a dict payload the handler forwards verbatim.

Common situations: Client code copying the OCR (det+rec) hub client but omitting the paths field default; JSON clients sending null instead of []; sending base64 strings where the module expects decoded ndarrays.

Related errors

• Environment Variable CUDA_VISIBLE_DEVICES is not set correct
• Environment Variable CUDA_VISIBLE_DEVICES is not set correct
• Unsupported model: {model!r}
• Unknown provider: {provider}
• Unsupported model: {normalized!r}. Supported models: {suppor

AI-assisted analysis of PaddlePaddle/PaddleOCR@2661c7c0ef (2026-08-14). Data as JSON: /api/errors/76971d3f83a87f76. Report an issue: GitHub.