keras-team/keras · error · ValueError

Invalid subset name: {subset}; expected "training" or "valid

Error message

Invalid subset name: {subset}; expected "training" or "validation".

What it means

ImageDataGenerator splits data via subset='training'/'validation' only. NumpyArrayIterator rejects any other subset string.

Source

Thrown at keras/src/legacy/preprocessing/image.py:581

            x_misc = []

        if y is not None and len(x) != len(y):
            raise ValueError(
                "`x` (images tensor) and `y` (labels) "
                "should have the same length. "
                f"Found: x.shape = {np.asarray(x).shape}, "
                f"y.shape = {np.asarray(y).shape}"
            )
        if sample_weight is not None and len(x) != len(sample_weight):
            raise ValueError(
                "`x` (images tensor) and `sample_weight` "
                "should have the same length. "
                f"Found: x.shape = {np.asarray(x).shape}, "
                f"sample_weight.shape = {np.asarray(sample_weight).shape}"
            )
        if subset is not None:
            if subset not in {"training", "validation"}:
                raise ValueError(
                    f"Invalid subset name: {subset}"
                    '; expected "training" or "validation".'
                )
            split_idx = int(len(x) * image_data_generator._validation_split)

            if (
                y is not None
                and not ignore_class_split
                and not np.array_equal(
                    np.unique(y[:split_idx]), np.unique(y[split_idx:])
                )
            ):
                raise ValueError(
                    "Training and validation subsets "
                    "have different number of classes after "
                    "the split. If your numpy arrays are "
                    "sorted by the label, you might want "
                    "to shuffle them."

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Use subset='training' or subset='validation' exactly
2. Pass validation_split=0.2 (0<split<1) to ImageDataGenerator when using subsets

Example fix

// before
gen.flow(x, y, subset='train')
// after
gen = ImageDataGenerator(validation_split=0.2)
gen.flow(x, y, subset='training')

assert subset in (None, 'training', 'validation')

def is_valid_subset(s): return s is None or s in {'training','validation'}

try: gen.flow(x, y, subset=s)
except ValueError as e: if 'Invalid subset' in str(e): s = 'training'

When it happens

Trigger: Calling gen.flow(x, y, subset='train'), subset='val', or subset='test' (also requires validation_split set on the generator).

Common situations: Copying tutorial code and renaming subsets; using subset without validation_split; Keras 2 vs 3 naming confusion.

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

• Unknown activation function '{activation}' cannot be seriali
• Could not interpret activation function identifier: {identif
• ConvNeXt does not support the `channels_first` image data fo
• If using `weights="imagenet"` with `include_top=True`, `clas
• The `weights` argument should be either `None` (random initi

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/4263661a99535d0f. Report an issue: GitHub.