From f28155105134b90fd86049c65478d307e0d8dbbc Mon Sep 17 00:00:00 2001 From: JSDurand Date: Sat, 28 Jan 2023 10:17:24 +0800 Subject: a prototype of an item derivation forest It seems to be complete now, but still awaits more tests to see where the errors are, which should be plenty, haha. --- ChangeLog | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'ChangeLog') diff --git a/ChangeLog b/ChangeLog index 44e241a..057ef5d 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,22 @@ +2023-01-28 Jean Sévère Durand + + * chain: Move the forest crate to the item module in the chain + crate. That crate used to provide a kind of forests. This turns + out to be an item derivation forest for the chain-rule machine. + So it seems appropriate to implement this in the chain crate + instead. + + * chain: Also the chain rule machine seems to be complete now. It + remains to test the machine to find the must-be errors. + + * graph: Now the doubly linked graphs do not exclude nodes with + duplicate edge sets. That is, previously, when adding a node + whose children set is equal to that of another existing node, the + graph just reuses that old node. Now this is not the case + anymore. The reason is that this only saves some minor memory + usage, at the cost of more expensive graph operations, and might + affect the time complixity, so I just remove this functionality. + 2023-01-22 Jean Sévère Durand * forest: Correctly clone nodes. -- cgit v1.2.3-18-g5258