No project description provided
Project description
Tabular Time Series
Summary
This repo was created as I did not find a function able to transform a time-series (1D) into a tabular format (X, y).
Usage
TimeSeriesGenerator
The docstring is as follows. Given a 1D array data = [0, 1, 2, 3, 4, 5, 6]
, generates X, y
following the parameters p
(autoregressive), s
(seasonal) and n
(lenght of y).
Therefore, it makes it possible to train a neural network (e.g.) that 2 autoregressive entries (e.g. p = 2
) and predicts the next two (n = 2
) using 2 (n = 2
) entries with lag 4 (s = 4
).
>> data = [0, 1, 2, 3, 4, 5, 6]
>> p, n = 2, 2
>> ts = TimeSeriesGenerator(data, p, n)
>> for X, y in ts:
... print(X, y)
[0, 1] [2, 3]
[1, 2] [3, 4]
[2, 3] [4, 5]
[3, 4] [5, 6]
>> p, n, s = 2, 2, 4
>> ts = TimeSeriesGenerator(data, p, n, s)
>> for X, y in ts:
... # both y have their respective seasonal entry
... print(data.index(y[0]) - data.index(X[0]) == s, data.index(y[1]) - data.index(X[1]) == s)
... print(X, y)
[0, 1, 2, 3] [4, 5]
[1, 2, 3, 4] [5, 6]
timeseries2df
Considering that many times a batch array is needed for training, timeseries2df
can be used to generate a pandas
DataFrame that will contain columns in the format:
>>> from tabular_time_series.tsdf import timeseries2df
>>> data = list(range(10))
>>> p, n, s = 2, 2, 4
>>> df = timeseries2df(data, p, n, s)
>>> df
y(ts4)_1 y(ts4)_2 y(t-1) y(t-0) y(t+1) y(t+2)
0 0 1 2 3 4 5
1 1 2 3 4 5 6
2 2 3 4 5 6 7
3 3 4 5 6 7 8
4 4 5 6 7 8 9
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for tabular-time-series-0.2.32.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 912757201c4781886c2b1fc879301cac0ab3da36caf8aa0d405b9f06731a82fa |
|
MD5 | ac02f01c6aa5d35728c0d45138240191 |
|
BLAKE2b-256 | e32bebfe4d635f4dfd19a1b163720702421fa87147ad08b85ff1140fbc88e96f |
Hashes for tabular_time_series-0.2.32-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | eeed2542d15a99ffd7f4caaec6bcae8e101ba8c628e3055ed522b7469d938ae9 |
|
MD5 | 40d5a2c1e8d2950bc75823437d4a8c8f |
|
BLAKE2b-256 | 7ba57bb86dd92d69ef1531402efa647393863fed6716fc750afbf5da1b5c885f |