mirror of
https://github.com/wassname/torch-neuralpointprocess.git
synced 2026-06-27 17:50:52 +08:00
wassname
32 lines
1.2 KiB
Markdown
32 lines
1.2 KiB
Markdown
Modified by wassname from the below:
|
|
|
|
Changes:
|
|
- [x] clamp weight with epsilon for stablity
|
|
- from `p.data *= (p.data>=0)`
|
|
- to `p.data = torch.clamp(p.data, min=eps)`
|
|
- [ ] try log t
|
|
- [x] try not mean as much in intensity layer
|
|
- [x] use pytorch lightning
|
|
- [x] fix potential data leak where label is fed to lsm
|
|
- [x] commit notebook with plots of val
|
|
|
|
|
|
# Fully Neural Network based Model for General Temporal Point Process(Neurips 2019,Takahiro Omi)
|
|
|
|
This code is pytorch version of implementation for Neural Temporal Point Process.
|
|
|
|
Temporal Point Process is mathematical model for capturing patterns of discrete event occurrences.
|
|
However, the traditional point process have limited expressivity by assumption.
|
|
For example, Poisson process assumes the independence of all events though it changes by time.
|
|
Other point process like Hawkes process does not assume the independence but the intensity function
|
|
of Hawkes process should be positive and have exponential decaying kernel. For these reason, the author
|
|
suggest the generalized version of point process by introducing neural network.
|
|
|
|
|
|
Reference
|
|
|
|
github : https://github.com/omitakahiro/NeuralNetworkPointProcess
|
|
|
|
|
|
|