tanitter

Hmm... it seems mysterious. I am sorry that I do not remember well about its implementation. What I remember is that I made some inferences for implementing __ofm_ because [SuperMemo...

Yeah, it was even not easy to grasp the overview of the algorithm when reading the website first time. I have created [a web service](http://slaypni.github.io/flashward/) for memorizing foreign words (particularly...

Oh, I did not read related papers. It could have helped me to implement the code if I knew it. Perhaps, it may be good to if there are some...

@ronggong Sorry for this late response. I do not understand what the local step size is. If it can be added to FastDTW algorithm, it may be better to start...

As you mentioned, there is an issue for installing cython version of fastdtw at this moment. There might be some problems in setup.py, however, I did not figure it out...

Thank you for the information. Maybe better to change `setup.py` to install cython version always. If someone knows how to do that without dropping portability (architecture compatibility), I would like...

I do not think it is valid to use DTW as a kernel function, although it could return a value.

I found a paper by googling. Although I did not read it, it might be informative. Wan, V., & Carmichael, J. (2005). Polynomial dynamic time warping kernel support vector machines...

@lvermue Thank you for the PR! The execution time improvement looks significant. Cloud you writes tests for the new functions? (Strikethroughed because of the following question)

@lvermue Is just writing something like this code insufficient? ```python import itertools from fastdtw import fastdtw from joblib import Parallel, delayed import numpy as np X = np.random.randint(1, 40, size=(100,...