## Describe the bug
I have observed a considerable decrease in policy performance after the recent PyTorch 2.5.0 update. The decrease in performance replicates when training with A2C, REINFORCE ...
## Describe the bug
In `rl4co/models/nn/graph/gcn.py`, the `GCNEncoder` class processes batched node embeddings without adjusting the edge_index for each graph in the batch. This leads to incorr...
## Description
See #225
## Motivation and Context
- [ ] I have raised an issue to propose this change ([required](https://github.com/ai4co/rl4co/issues) for new features and bug fixes)
...
## Motivation
The `ATSPGenerator._generate` function seems to have high complexity in both time and space.
## Solution
I have tried to implement Floyd-Warshall algorithm as an alternative,...
Hello, I can't solve the instances in cvrplib with the trained model.
Traceback (most recent call last):
File "/NCO/RL4CO/examples/test_cvrplib_demo.py", line 91, in <module>
out = polic...