I am trying to take a look at this huggingface model in their documentation using tensorboard:
As you can see, the model takes in a lot of inputs:
outputs = model( … past_values=batch[“past_values”], … past_time_features=batch[“past_time_features”], … past_observed_mask=batch[“past_observed_mask”], … static_categorical_features=batch[“static_categorical_features”], … static_real_features=batch[“static_real_features”], … future_values=batch[“future_values”], … future_time_features=batch[“future_time_features”])
So I try to add a model to tensorboard like this:
writer.add_graph(model, (batch[“past_values”],batch[“past_time_features”],
batch[“past_observed_mask”],
batch[“static_categorical_features”],
batch[“static_real_features”],
batch[“future_values”],
batch[“future_time_features”],
))
But I get this error: :Dictionary inputs to traced functions must have consistent type. Found Tensor and Tuple[Tensor, Tensor, Tensor]
Any ideas on the proper syntax?