-
My use case is that I would like to use graph_nets to perform logical reasoning on the output of some prior batched TF operations. Let me present an example:
``` python
B,W,H,C = 10, 200,200,3
da…
-
OTP1 support loading multiple graphs (routers) and this is part of the URLs. In most deployments only one router is used (the default router).
This make the code complicated; hence increase the…
-
Hi @kramsretlow! Nice work!
Would you be interested in integrating your solver within the GeoStats.jl framework? It is a perfect fit in my opinion, specially regarding the spatial examples you gave…
-
Some graphnodes operates on any type of Spatial nodes, but their output is always of type NODE, which means we lose information after going through them.
![image](https://user-images.githubusercont…
-
When I try to filter on properties (e..g., cloud_cover) in the load_collection process_graph, I get an exception (translate.py line 299 in adjust_from_parameters):
`ValueError: 'from_parameter' refer…
-
Here's a reading and discussion exercise we'll be running for the next few days. Please complete your posts by next Monday, 2/24. Choose at least one of the following two articles by Steve Ramsay to r…
-
So my intuitive guess for how the MNIST dataset worked would be classifying the graph of just the positive pixels into 0-9. The example being a 2.
![image](https://user-images.githubusercontent.com/1…
-
We wanted at Plannerstack to update to the latest version of Master. We however found out that the performance of the latest Master has degraded quite a bit, in some cases taking 4 times longer to ret…
-
Hey again Marius! I have another question about chemlambda. First, some background on the situation:
I'm currently implementing the moves for chemlambda-hask. In my implementation of chemlambda, a gr…
-
L4T7_KEY name suggests it is using 4 spatial and 7 temporal layers.
That is not entirely true. It is using 4 spatial ids and 7 temporal ids to describes 3 spatial and 3 temporal layers.
Extra ids ar…