qubo-nn
covalent
qubo-nn | covalent | |
---|---|---|
3 | 4 | |
40 | 698 | |
- | 2.5% | |
0.0 | 8.6 | |
over 2 years ago | 9 days ago | |
Python | Python | |
MIT License | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
qubo-nn
- Reverse-Engineering QUBO Matrices
-
How do you handle imbalanced datasets?
Here is an example.
-
I accidentally tried to train 1 Quadrillion parameters - and a story of structural bias
In the course of that project (see here: https://github.com/instance01/qubo-nn) the second part turned out to be quite tough, likely due to the high dimensionality (4096 input size). In the post above I go through a mistake I made in which I assumed underfitting whereas in reality my data had structural bias (randomness that cannot be learnt by a neural network).
covalent
-
Remote execution of code
Pretty interesting request, if SSH is not used, i would try using something like dask which uses tcp to connect and execute assuming your workers are in another machine.I also think something like covalent can be used to extend your own custom plugin in their ecosystem to connect how you want. We have a very custom private plugin written on top of covalent's to have a custom protocol to connect our central on-prem GPU machines to our local laptops that is rpc based, mostly for high performance as well as some mandate security from where the GPU machines are. Once done it is pretty much something like
-
Prefect alternatives meant for Slurm (HPC)
Does anyone here have any suggestions of alternatives tailored for Slurm on HPC? I know Covalent is one option, but I'm curious about others as well. Ideally the platform should be Pythonic, have a GUI, and be reasonably active/well-maintained.
- Show HN: Covalent – distributed computing for ML, HPC and Quantum (open source)
-
Your strategies for offloading computation
Came across this new tool exactly for this - https://github.com/AgnostiqHQ/covalent
What are some alternatives?
Cirq - A python framework for creating, editing, and invoking Noisy Intermediate Scale Quantum (NISQ) circuits.
SmartSim - SmartSim Infrastructure Library.
PyTorch-VAE - A Collection of Variational Autoencoders (VAE) in PyTorch.
kestra - Infinitely scalable, event-driven, language-agnostic orchestration and scheduling platform to manage millions of workflows declaratively in code.
quantum-nearest-classifier - Cirq implementation of Nearest Centroid Classification on a Trapped Ion Quantum Computer (Johri et al., 2020)
cadence-python - Python framework for Cadence Workflow Service
qsearch - A compiler for quantum computers based on A* and numerical optimization.
mlnotify - 🔔 No need to keep checking your training - just one import line and you'll know the second it's done.
dagster - An orchestration platform for the development, production, and observation of data assets.
bytehub - ByteHub: making feature stores simple
streetdensityai - This YoloV5 based model is fit to detect people and different types of land vehicles, and displaying their density on a fitted map, according to their coordinates and detected labels.
hera - Hera is an Argo Python SDK. Hera aims to make construction and submission of various Argo Project resources easy and accessible to everyone! Hera abstracts away low-level setup details while still maintaining a consistent vocabulary with Argo. ⭐️ Remember to star!