mbchang comments

Results 17 comments of


                                            mbchang

Tensorflow

Hi, the Neural Physics Engine is very similar in implementation to the Interaction Network and Relation Network. While the NPE is implemented in Torch, there are tensorflow implementations of the...

Hi, while the Social LSTM does preserve some spatial information, it loses spatial information through the social pooling operation (for neighbors within a certain spatial distance). Similarly, the No-Pairwise baseline...

Cuda support for affine transforms

It would be helpful to integrate cuda support more deeply into torchsample. For example, in torchsample/transforms/affine_transforms.py, instead of ``` zoom_matrix = th.FloatTensor([[zx, 0, 0], [0, zy, 0], [0, 0, 1]])...

Problem: Index tensor must have same dimensions as output tensor

I'm having the same problem too.

Adds Reflexion (solves #2316)

cc @noahshinn024

Adding a Character Chat Agent

Thanks @treppers for the PR! This kind of agent looks like it could make writing dialogue agents much more streamlined. - It looks like you have made a few changes...

Adding a Character Chat Agent

Thanks for the update @treppers! Curious to see what you find in the compare-and-contrast.

Agent simulations

This post is about `agent_simulations_v3`, which improves upon `agent_simulations_v2` by using a generic `RoundRobinSimulator` rather than a CAMEL-specific simulator. If all we want is a simulator for two agents, then...

fix: assign current_time to datetime.now() if current_time is None

`current_time = kwargs.get("current_time")` is not sufficient, because even though the variable `current_time` exists as a `kwarg`, the value passed for the `current_time` is `None`. This causes `doc.metadata["last_accessed_at"] = None` and...

fix: assign current_time to datetime.now() if current_time is None

oh yeah makes sense; I've updated it