Maximilian Sieb

Hello, I was wondering whether it is possible to use a feature or signal for reward calculation without explicitly having the trajectory optimizer using it as part of the state...

I was wondering if there is a quick way of interfacing the library with Baxter in Gazebo by changing src code within gps_agent_pkg and the corresponding launch files?

point_cloud_classification.ipynb: TypeError: initialize() got an unexpected keyword argument 'device'

2

In the example notebook, the dataset class does not accept a `device` argument. How does one cast the loader to `cuda` in this example?: ``` val_loader = DataLoader(ModelNet(modelnet_path, categories=categories, split='test',transform=transform,...

Running the cell ``` engine = ClassificationEngine(PointNet(num_classes=len(categories)), train_loader, val_loader) engine.fit() ``` gives ``` 224 225 for idx, batch in enumerate(self.train_loader): --> 226 loss, accuracy = self.train_batch(batch) 227 self.optimize_batch(loss) 228 self.print_train_stats()...

`initialize_agent` does not work with `return_intermediate_steps=True`

10

E.g. running ```python from langchain.agents import load_tools from langchain.agents import initialize_agent from langchain.llms import OpenAI llm = OpenAI(temperature=0) tools = load_tools(["llm-math"], llm=llm) agent = initialize_agent(tools, llm, agent="zero-shot-react-description", verbose=True, return_intermediate_steps=True) agent.run("What...