-
Hi Victor, do you need a js visualization for the formality-core? I just have one which can do interaction combinators [1] and it seems it is easy to adapt for your needs. As I am not competent in the…
-
#### Current Behaviour
Currently the HF net routing code (spatial look-up to avoid putting large route trees repeatedly on the heap) is only active for non-timing critical connections (e.g. critica…
-
[SchNet](https://arxiv.org/abs/1706.08566)-like models have significant advantages in that they use atom environment embeddings instead of separate pairs of networks to compute energy contributions. W…
-
Hiya! I was attempting to help Joel and Genny locate some sources of lag on their server for Genny's latest modpack. While I did have access to samples, I didn't have direct access to the server, so I…
-
I'll create this new issue to track new ideas and channel early questions.
-
I was playing around with thaumcraft and after I did _something_, I started having this issue with AE machines (terminals, interfaces, buses, etc). My **crafting** terminal looked like this https://gy…
-
How do I make this program work?
import Base#
laver(a:Nat, b:Nat) : Nat
if a==0 then b else if b==3 then a-1 else laver(laver(a,b+1),a-1)
main : Nat
laver(3,1)
When run with fm -d ma…
tromp updated
4 years ago
-
There are several reasons for noise in results across different runs. It would be good if we are able to control the randomness in the experiment with a single random seed. In case this is not possibl…
-
Hello,
I was checking policy gradient algorithm on HLE.
Firstly, I think there is something wrong with how Hanabi handles rewards (or I am missing something). I ran this simple code (similar to PG f…
-
```
graph_network = modules.InteractionNetwork(
edge_model_fn=lambda: snt.nets.MLP(output_size=depth, dropout_rate=0.2),
node_model_fn=lambda: snt.nets.MLP(output_size=depth, dropou…