conditional-flow-matching
conditional-flow-matching copied to clipboard
atong01
TorchCFM: a Conditional Flow Matching library
Thanks very much for this code base, it's been a great way to learn about flow matching. I have a question regarding conditional generation with OT-CFM. When testing different FM...
In the current implementation, tensors are moved to numpy + CPU before calling the optimal transport solver, see e.g. https://github.com/atong01/conditional-flow-matching/blob/main/torchcfm/optimal_transport.py#L88. Since version 0.8, the POT package supports backends beyond numpy,...
Hi, how to sample from batch OT if batch size used for compute OT is large than training batch? For example, I use batch size 128 to compute batch OT,...
I tried to change the unconditional flow matching algorithm for CIFAR10 to a conditional one, like with MNIST, but I failed and encountered unknown errors. How should I implement conditional...
- added more tests for running forward functions of models - test which expands Unet to take floating point conditions (or anything else) - edited Unet to accept custom embedding...
Dear all, thanks again for the library. So far, I haven't found anything equivalent to exploit the conditional flow matching paradigm. To learn the library, I started from the example...
Hi great work ! will you in the future release also code for cifar 100 ? or maybe image-net ?
sry but is `https://github.com/atong01/conditional-flow-matching/blob/master/examples/notebooks/model-comparison-plotting.ipynb` unavailable now?
Hi everyone, Thanks for your contribution in this package! I am quite new in the Flow Matching field, and for my research I believe that I could utilize OT-CFM to...
