Rene Pickhardt
Rene Pickhardt
Right now the `UncertaintyChannel` is the place where we mainly provide two features for our cost function. It is pretty obvious that some feature representing physical distance / latency of...
@sebulino rightfully pointed out in out of band communication that a dict for payments that is indexed by an incremental integer is just a list. Yet I introduced such a...
This is an example showing how unit tests could be added which would be a first step towards fixing #8 . However I am not sure if this is really...
As realized through discussions with @sebulino in #15 the current lib does only decide if a payment can be delivered or not. In many cases it would be preferable that...
As discussed with @sebulino the payment session currently computes a lot of statistics and depicts them to the user via the command line. However it is pretty pointless if one...
Currently I am trying to figure out if the API of the package makes sense. Once this is done we should include test cases for all classes and fix the...
To test on mainnet one only has to provide a wrapper to a lightning node implementation as the oracle. I wonder if the lib should provide a Mainnet Wrapper to...
Currently the following things are being ignored when adding edges to the min cost flow solver and when dissecting the result into paths: * `max htlc` that is possible per...
This is delicate as it should also account for other engineering challenges (HTLC limits, channel reserves, stuck payments) as people might include this quickly into their Applications
In https://github.com/renepickhardt/mpp-splitter/issues/12 we already used better pruning. The current package still has the ad-hoc reliability based pruning method. This needs to be changed to cost based pruning and automate this...