tensorflow/models · error · ValueError

predict requires at least one image.

Error message

predict requires at least one image.

What it means

Error "predict requires at least one image." thrown in tensorflow/models.

Source

Thrown at official/projects/waste_identification_ml/Deploy/pet_grading_cloud_deployment/pet_grade_classifier.py:226

    """Classifies a list of PIL images in a single forward pass.

    Args:
        pil_images: Sequence of RGB PIL images (length >= 1).

    Returns:
        One prediction dict per input image, in the same order::

            {
                "predicted_class": str,
                "predicted_probability": float,  # 0-100
                "all_probabilities": dict[str, float],
            }

    Raises:
        ValueError: If ``pil_images`` is empty.
    """
    if not pil_images:
      raise ValueError("predict requires at least one image.")

    batch = torch.stack([self._eval_transform(img) for img in pil_images]).to(
        self.device
    )

    logits = self.model(batch)
    probabilities = torch_functional.softmax(logits, dim=1).cpu().numpy()

    predictions = []
    for probability_row in probabilities:
      predicted_index = int(np.argmax(probability_row))
      predictions.append({
          "predicted_class": self.class_names[predicted_index],
          "predicted_probability": float(
              probability_row[predicted_index] * 100.0
          ),
          "all_probabilities": {
              name: float(p * 100.0)

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/waste_identification_ml/Deploy/pet_grading_cloud_deployment/pet_grade_classifier.py:226 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tensorflow/models@e006f5f0d5 (2026-08-24). Data as JSON: /api/errors/3a904d2df4fa85c4. Report an issue: GitHub.