Xie Chong

If you have interests in adding your own photo album, you could take a look at the plugin that I am using. [https://github.com/xcTorres/xcTorres.github.io](https://github.com/xcTorres/xcTorres.github.io)

Correction doesn't prioritize bigram.

3

There are **sengkerang** and **selatan** in gram dictionary, the frequency are 500. And there is **tangkerang selatan** in bigram dictionary, the frequency is 1200. When correcting the address "Jalan Surabaya...

When I doing the loading test, it reports Too many open files error. Is the error from express or osrm-backend?

when I use nanocube-binning-csv script,it always occur this problem "Segmentation fault (core dumped)"

2

nanocube-binning-csv --sep=',' --latcol='pickup_latitude' --loncol='pickup_longitude' --timecol='tpep_pickup_datetime' --timebinsize='1D' ./yellow_tripdata_2016-01.csv > ./taxi.dmp Segmentation fault (core dumped)

Cost Problem

4

Thank you for your library, it is very fast. But when I calculate this matrix the result seems like not true. -625.000 2187.500 -156.25 1000000 -2500.000 1000000 -2500.00 -2500.0 -1015.625...

hash trick

1

你好，我是初学ffm，我能问个问题么，在这个编码过程中bin=10的五次方，那么如果有个数据量用户有500万个，那么按道理特征的index是不是不仅仅在10的五次方里面。这是我的一个疑问哈