Troubleshoot

Backend Not Available

If you select a backend that is not installed, the library raises an import error.

from hdmrlib import set_backend

set_backend("torch")

Install the missing backend and try again.

pip install hdmrlib[torch]

The same applies to TensorFlow.

pip install hdmrlib[tensorflow]

If the backend name is not recognized, the library raises a value error.

set_backend("unknown_backend")

Use one of the supported backend names:

If no supported backend is available, the library raises a runtime error when it tries to create a backend instance.

Make sure that at least one supported backend is installed.

If the input tensor contains singleton dimensions, the backend implementation squeezes them and prints a warning.

For example, an input with shape (1, 10, 10) is converted to (10, 10) internally.

Check the tensor shape before running the decomposition if the exact dimensional structure matters.

If you use supports="custom" without providing custom_supports, the library raises a value error.

from hdmrlib import EMPR

empr = EMPR(X, order=2, supports="custom")

Provide one support vector per input dimension.

If the number of custom supports does not match the number of tensor dimensions, the library raises a value error.

custom_supports = [np.ones((10, 1))]

For a two-dimensional tensor, provide two support vectors.

If you use weight="custom" in HDMR without providing custom_weights, the library raises a value error.

from hdmrlib import HDMR

hdmr = HDMR(X, order=2, weight="custom")

Provide one weight vector per input dimension.

If the number of custom weights does not match the number of tensor dimensions, the library raises a value error.

For a three-dimensional tensor, provide three weight vectors.

components(elements=[...]) expects existing component keys.

components = empr.components(elements=["g_1", "g_2", "g_1,2"])

Check available keys first:

print(empr.components().keys())

Use the returned keys exactly as they appear.