Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/fridex/ml_kiwi
kiwi.com's machine learning weekend
https://github.com/fridex/ml_kiwi
Last synced: about 1 month ago
JSON representation
kiwi.com's machine learning weekend
- Host: GitHub
- URL: https://github.com/fridex/ml_kiwi
- Owner: fridex
- Created: 2017-09-26T14:49:18.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2022-12-08T00:35:40.000Z (about 2 years ago)
- Last Synced: 2024-04-16T07:20:02.418Z (8 months ago)
- Language: Python
- Size: 159 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# kiwi.com Machine Learning Weekend
See description at [https://www.kiwi.com/mlweekend/](https://kiwi.com/mlweekend)
## Installation
This project is not available on PyPI, so to install it (directly master) issue:
```bash
$ pip3 install git+https://github.com/fridex/ml_kiwi
```
Once installation succeeds, run:
```
$ ml-kiwi --help
Usage: ml-kiwi [OPTIONS]
Compute degree of polynomial for kiwi.com's weekend machine learning
session.
Options:
-i, --input PATH Input file to be used - if omitted, remote API is
called to retrieve values.
-o, --output PATH Output file to be used - if omitted, defaults to
stdout).
--output-image PATH Output file where interpolation image should be
stored.
--max-degree INTEGER Maximal degree of fitting polynomial (default: 100).
--min-degree INTEGER Minimal degree of fitting polynomial (default: 1)
--x-from FLOAT Value on x axis to start with data gathering from API
(default: -100.000000).
--x-to FLOAT Value on x axis that shouldn't be reached when
gathering data from API (default: 100.000000).
--x-step FLOAT Step on x axis to start with (default: 0.050000).
-v, --verbose Turn on debug messages.
--no-pretty Turn off pretty formatted output.
--no-show-plot Do not show plot results.
--help Show this message and exit.
```
## Troubleshooting
I'm getting the following traceback when running `ml-kiwi` executable:
```
[root@host /]# ml-kiwi
Traceback (most recent call last):
File "/usr/bin/ml-kiwi", line 10, in
from ml_kiwi import Defaults, DataSource, do_interpolate, do_plot, json_dump
File "/usr/lib/python3.5/site-packages/ml_kiwi/__init__.py", line 12, in
from .lib import DataSource
File "/usr/lib/python3.5/site-packages/ml_kiwi/lib.py", line 11, in
import matplotlib.pyplot as plt
File "/usr/lib64/python3.5/site-packages/matplotlib/pyplot.py", line 115, in
_backend_mod, new_figure_manager, draw_if_interactive, _show = pylab_setup()
File "/usr/lib64/python3.5/site-packages/matplotlib/backends/__init__.py", line 32, in pylab_setup
globals(),locals(),[backend_name],0)
File "/usr/lib64/python3.5/site-packages/matplotlib/backends/backend_tkagg.py", line 6, in
from six.moves import tkinter as Tk
File "/usr/lib/python3.5/site-packages/six.py", line 92, in __get__
result = self._resolve()
File "/usr/lib/python3.5/site-packages/six.py", line 115, in _resolve
return _import_module(self.mod)
File "/usr/lib/python3.5/site-packages/six.py", line 82, in _import_module
__import__(name)
ImportError: No module named 'tkinter'
```
You need to install Python3 compliant version of tkinter (on Fedora):
```bash
$ dnf install python3-tkinter
$ # alternatively for deb based distributions
$ # apt-get install python3-tkinter
```
# Output
The script will by default query endpoint provided by kiwi.com and check for y values for the given x (x and step on x axis adjustable via CLI arguments). These values are used for interpolating polynomial (one can explicitly specify polynomial degree via CLI arguments). The result is plotted and, if requested, stored to output file.
The resulting JSON can be re-used for computing interpolating polynomial again, point to it using `--input` parameter. The API endpoint will not be queried in that case, but collected data would be reused.
## Example output
You can find examples of kiwi.com input in `results` directory. The script was run with default values as follows:
```bash
$ ml-kiwi -o results/output.json --output-image results/output.png
```
## License
MIT