Description

Define transform_observation_spec function in _R3MNet class

1. if del_keys is True, we delete all items in the input observation_spec
2. We iterate the keys_out, for each output key, we generate NdUnboundedContinuousTensorSpec objects with corresponding ResNet output dimension as shape

Add test cases

One test case TestR3MNet.test_r3mnet_transform_observation_spec is added. In the test, we set up these parameters

• models: resnet 18/34/50
• del_keys: True or False
• keys_in: [["next_pixels"], ["next_pixels_1", "next_pixels_2", "next_pixels_3"]]
• keys_out: [["next_r3m_vec"], ["next_r3m_vec_1", "next_r3m_vec_2", "next_r3m_vec_3"]]

Then we compare the new observation_spec with the expected ones.

Motivation and Context

close #407

Types of changes

What types of changes does your code introduce? Remove all that do not apply:

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds core functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation (update in the documentation)
• Example (update in the folder of examples)

Checklist

• I have read the CONTRIBUTION guide (required)
• My change requires a change to the documentation.
• I have updated the tests accordingly (required for a bug fix or a new feature).
• I have updated the documentation accordingly.