Tensorflow - ValueError: Failed to convert a NumPy array to a Tensor (Unsupported object type float)

Background

You're trying to train a model with a list of lists, each containing 1000 floats, but encounter the error "Failed to convert a NumPy array to a Tensor (Unsupported object type float)."

Cause and Solution

Tensorflow requires input data to be in the form of tensors, not lists. In this case, the error is caused by you passing lists as input to your model. To resolve this, convert your training data to a NumPy array using the following code:

x_train = np.asarray(x_train).astype('float32')

Ensure that your data is properly formatted, addressing issues such as categoricals, NaNs, and strings. Additionally, verify that your model's input and output shapes align with the data dimensions it expects.

For LSTM models, the expected data dimensions are (batch_size, timesteps, features). You can use the following code to print the shapes of your model's inputs and outputs:

[print(i.shape, i.dtype) for i in model.inputs]
[print(o.shape, o.dtype) for o in model.outputs]

To debug data issues, print the shapes of both input and output data to confirm they conform to the expected format. Additionally, consider using an IDE like Spyder that supports cell-based execution to facilitate debugging.