-
Hi
I ran LAG with MAPPO & HAPPO but the eval plot is looking weird. Can you please share the LAG hyperparameters for MAPPO & HAPPO?
This is regarding fig. 10(c) in the paper: MAXIMUM ENTROPY HETER…
-
# Describe the bug
A clear and concise description of what the bug is.
**Please share your notebook link so that we can reproduce the error**
I have a model trained for the Actor Critic model i…
-
from safepo.common.lagrange import Lagrange
nu = 1.0
nu_lr = 0.1
ep_cost = 35
lagrange = Lagrange(
cost_limit=25.0,
lagrangian_multiplier_init=1.0,
lagrangian_multiplier_lr=0.1,
…
-
# WHY
The types of applications you can build with just discrete actions are limited. Supporting continuous actions allows for building more powerful apps. This makes Spice.ai more useful for many …
-
Hello, I met a value error when running the code directly. I extracted the relevant code in Section 1 and got the error in Section 2. It works after I add an extra code `actor_params = freeze(actor_pa…
-
Hi,thank you for providing this good work. But i still have a little question. In the paper, history h_t is defined as the historical sequence of observations and actions like (o_1,a_1,……o_T,a_T). How…
-
**Describe the bug**
First of all, I am not sure whether it is a bug or not.
I was training my own model with DDP enabled which ran into this error. Then I took examples/distributed_offline_tra…
-
The `TorchBeastEncoder` is used as the default reward encoder in Motif. However, in the Actor-Critic, the default encoder is `NLEMainEncoder`. We noticed a significant performance degradation when we …
-
Find a way to add "throughput" (or something similar) to logging in `sumo_generate_learning_curve.py` This metric can either be tracked over the course of an entire episode (in which case the average …
-
#108 is going to be an important place to start.
Twitter is the only context where I do not currently see a need for increased emphasis on self-curation at this time considering how that all appearan…