An intuitive library to add plotting functionality to scikitlearn objects.
Project description
# Welcome to Scikitplot
[![PyPI version](https://badge.fury.io/py/scikitplot.svg)](https://badge.fury.io/py/scikitplot)
[![license](https://img.shields.io/github/license/mashape/apistatus.svg)]()
[![Build Status](https://travisci.org/reiinakano/scikitplot.svg?branch=master)](https://travisci.org/reiinakano/scikitplot)
[![PyPI](https://img.shields.io/pypi/pyversions/scikitplot.svg)]()
[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.293191.svg)](https://doi.org/10.5281/zenodo.293191)
### Single line functions for detailed visualizations
### The quickest and easiest way to go from analysis...
![roc_curves](docs/_static/readme_collage.jpg)
### ...to this.
Scikitplot is the result of an unartistic data scientist's dreadful realization that *visualization is one of the most crucial components in the data science process, not just a mere afterthought*.
Gaining insights is simply a lot easier when you're looking at a colored heatmap of a confusion matrix complete with class labels rather than a singleline dump of numbers enclosed in brackets. Besides, if you ever need to present your results to someone (virtually any time anybody hires you to do data science), you show them visualizations, not a bunch of numbers in Excel.
That said, there are a number of visualizations that frequently pop up in machine learning. Scikitplot is a humble attempt to provide aestheticallychallenged programmers (such as myself) the opportunity to generate quick and beautiful graphs and plots with as little boilerplate as possible.
## Okay then, prove it. Show us an example.
Say we use Naive Bayes in multiclass classification and decide we want to visualize the results of a common classification metric, the Area under the Receiver Operating Characteristic curve. Since the ROC is only valid in binary classification, we want to show the respective ROC of each class if it were the positive class. As an added bonus, let's show the microaveraged and macroaveraged curve in the plot as well.
Let's use scikitplot with the sample digits dataset from scikitlearn.
```python
# The usual traintest split mumbojumbo
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB
X, y = load_digits(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)
nb = GaussianNB()
nb.fit(X_train, y_train)
predicted_probas = nb.predict_proba(X_test)
# The magic happens here
import matplotlib.pyplot as plt
import scikitplot as skplt
skplt.metrics.plot_roc(y_test, predicted_probas)
plt.show()
```
![roc_curves](examples/roc_curves.png)
Pretty.
And... That's it. Encaptured in that small example is the entire philosophy of Scikitplot: **single line functions for detailed visualization**. You simply browse the plots available in the documentation, and call the function with the necessary arguments. Scikitplot tries to stay out of your way as much as possible. No unnecessary bells and whistles. And when you *do* need the bells and whistles, each function offers a myriad of parameters for customizing various elements in your plots.
Finally, compare and [view the nonscikitplot way of plotting the multiclass ROC curve](http://scikitlearn.org/stable/auto_examples/model_selection/plot_roc.html). Which one would you rather do?
## Maximum flexibility. Compatibility with nonscikitlearn objects.
Although Scikitplot is loosely based around the scikitlearn interface, you don't actually need Scikitlearn objects to use the available functions. As long as you provide the functions what they're asking for, they'll happily draw the plots for you.
Here's a quick example to generate the precisionrecall curves of a Keras classifier on a sample dataset.
```python
# Import what's needed for the Functions API
import matplotlib.pyplot as plt
import scikitplot as skplt
# This is a Keras classifier. We'll generate probabilities on the test set.
keras_clf.fit(X_train, y_train, batch_size=64, nb_epoch=10, verbose=2)
probas = keras_clf.predict_proba(X_test, batch_size=64)
# Now plot.
skplt.metrics.plot_precision_recall_curve(y_test, probas)
plt.show()
```
![p_r_curves](examples/p_r_curves.png)
You can see clearly here that `skplt.metrics.plot_precision_recall_curve` needs only the ground truth yvalues and the predicted probabilities to generate the plot. This lets you use *anything* you want as the classifier, from Keras NNs to NLTK Naive Bayes to that groundbreaking classifier algorithm you just wrote.
The possibilities are endless.
## Installation
Installation is simple! First, make sure you have the dependencies [Scikitlearn](http://scikitlearn.org) and [Matplotlib](http://matplotlib.org/) installed.
Then just run:
```bash
pip install scikitplot
```
Or if you want the latest development version, clone this repo and run
```bash
python setup.py install
```
at the root folder.
If using conda, you can install Scikitplot by running:
```bash
conda install c condaforge scikitplot
```
## Documentation and Examples
Explore the full features of Scikitplot.
You can find detailed documentation [here](http://scikitplot.readthedocs.io).
Examples are found in the [examples folder of this repo](examples/).
## Contributing to Scikitplot
Reporting a bug? Suggesting a feature? Want to add your own plot to the library? Visit our [contributor guidelines](CONTRIBUTING.md).
## Citing Scikitplot
Are you using Scikitplot in an academic paper? You should be! Reviewers love eye candy.
If so, please consider citing Scikitplot with DOI [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.293191.svg)](https://doi.org/10.5281/zenodo.293191)
#### APA
> Reiichiro Nakano. (2018). reiinakano/scikitplot: 0.3.7 [Data set]. Zenodo. http://doi.org/10.5281/zenodo.293191
#### IEEE
> [1]Reiichiro Nakano, “reiinakano/scikitplot: 0.3.7”. Zenodo, 19Feb2017.
#### ACM
> [1]Reiichiro Nakano 2018. reiinakano/scikitplot: 0.3.7. Zenodo.
Happy plotting!
[![PyPI version](https://badge.fury.io/py/scikitplot.svg)](https://badge.fury.io/py/scikitplot)
[![license](https://img.shields.io/github/license/mashape/apistatus.svg)]()
[![Build Status](https://travisci.org/reiinakano/scikitplot.svg?branch=master)](https://travisci.org/reiinakano/scikitplot)
[![PyPI](https://img.shields.io/pypi/pyversions/scikitplot.svg)]()
[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.293191.svg)](https://doi.org/10.5281/zenodo.293191)
### Single line functions for detailed visualizations
### The quickest and easiest way to go from analysis...
![roc_curves](docs/_static/readme_collage.jpg)
### ...to this.
Scikitplot is the result of an unartistic data scientist's dreadful realization that *visualization is one of the most crucial components in the data science process, not just a mere afterthought*.
Gaining insights is simply a lot easier when you're looking at a colored heatmap of a confusion matrix complete with class labels rather than a singleline dump of numbers enclosed in brackets. Besides, if you ever need to present your results to someone (virtually any time anybody hires you to do data science), you show them visualizations, not a bunch of numbers in Excel.
That said, there are a number of visualizations that frequently pop up in machine learning. Scikitplot is a humble attempt to provide aestheticallychallenged programmers (such as myself) the opportunity to generate quick and beautiful graphs and plots with as little boilerplate as possible.
## Okay then, prove it. Show us an example.
Say we use Naive Bayes in multiclass classification and decide we want to visualize the results of a common classification metric, the Area under the Receiver Operating Characteristic curve. Since the ROC is only valid in binary classification, we want to show the respective ROC of each class if it were the positive class. As an added bonus, let's show the microaveraged and macroaveraged curve in the plot as well.
Let's use scikitplot with the sample digits dataset from scikitlearn.
```python
# The usual traintest split mumbojumbo
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import GaussianNB
X, y = load_digits(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)
nb = GaussianNB()
nb.fit(X_train, y_train)
predicted_probas = nb.predict_proba(X_test)
# The magic happens here
import matplotlib.pyplot as plt
import scikitplot as skplt
skplt.metrics.plot_roc(y_test, predicted_probas)
plt.show()
```
![roc_curves](examples/roc_curves.png)
Pretty.
And... That's it. Encaptured in that small example is the entire philosophy of Scikitplot: **single line functions for detailed visualization**. You simply browse the plots available in the documentation, and call the function with the necessary arguments. Scikitplot tries to stay out of your way as much as possible. No unnecessary bells and whistles. And when you *do* need the bells and whistles, each function offers a myriad of parameters for customizing various elements in your plots.
Finally, compare and [view the nonscikitplot way of plotting the multiclass ROC curve](http://scikitlearn.org/stable/auto_examples/model_selection/plot_roc.html). Which one would you rather do?
## Maximum flexibility. Compatibility with nonscikitlearn objects.
Although Scikitplot is loosely based around the scikitlearn interface, you don't actually need Scikitlearn objects to use the available functions. As long as you provide the functions what they're asking for, they'll happily draw the plots for you.
Here's a quick example to generate the precisionrecall curves of a Keras classifier on a sample dataset.
```python
# Import what's needed for the Functions API
import matplotlib.pyplot as plt
import scikitplot as skplt
# This is a Keras classifier. We'll generate probabilities on the test set.
keras_clf.fit(X_train, y_train, batch_size=64, nb_epoch=10, verbose=2)
probas = keras_clf.predict_proba(X_test, batch_size=64)
# Now plot.
skplt.metrics.plot_precision_recall_curve(y_test, probas)
plt.show()
```
![p_r_curves](examples/p_r_curves.png)
You can see clearly here that `skplt.metrics.plot_precision_recall_curve` needs only the ground truth yvalues and the predicted probabilities to generate the plot. This lets you use *anything* you want as the classifier, from Keras NNs to NLTK Naive Bayes to that groundbreaking classifier algorithm you just wrote.
The possibilities are endless.
## Installation
Installation is simple! First, make sure you have the dependencies [Scikitlearn](http://scikitlearn.org) and [Matplotlib](http://matplotlib.org/) installed.
Then just run:
```bash
pip install scikitplot
```
Or if you want the latest development version, clone this repo and run
```bash
python setup.py install
```
at the root folder.
If using conda, you can install Scikitplot by running:
```bash
conda install c condaforge scikitplot
```
## Documentation and Examples
Explore the full features of Scikitplot.
You can find detailed documentation [here](http://scikitplot.readthedocs.io).
Examples are found in the [examples folder of this repo](examples/).
## Contributing to Scikitplot
Reporting a bug? Suggesting a feature? Want to add your own plot to the library? Visit our [contributor guidelines](CONTRIBUTING.md).
## Citing Scikitplot
Are you using Scikitplot in an academic paper? You should be! Reviewers love eye candy.
If so, please consider citing Scikitplot with DOI [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.293191.svg)](https://doi.org/10.5281/zenodo.293191)
#### APA
> Reiichiro Nakano. (2018). reiinakano/scikitplot: 0.3.7 [Data set]. Zenodo. http://doi.org/10.5281/zenodo.293191
#### IEEE
> [1]Reiichiro Nakano, “reiinakano/scikitplot: 0.3.7”. Zenodo, 19Feb2017.
#### ACM
> [1]Reiichiro Nakano 2018. reiinakano/scikitplot: 0.3.7. Zenodo.
Happy plotting!
Project details
Release history Release notifications
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Filename, size  File type  Python version  Upload date  Hashes 

Filename, size scikit_plot0.3.7py3noneany.whl (33.7 kB)  File type Wheel  Python version py3  Upload date  Hashes View hashes 
Filename, size scikitplot0.3.7.tar.gz (28.6 kB)  File type Source  Python version None  Upload date  Hashes View hashes 
Close
Hashes for scikit_plot0.3.7py3noneany.whl
Algorithm  Hash digest  

SHA256  6b3d529800b32a899c54dc5761a93c63cbff482b1889a4afee57dd219f3ef0c3 

MD5  2e97f0b5e8fb7351555badc0ee83740d 

BLAKE2256  7c4732520e259340c140a4ad27c1b97050dd3254fdc517b1d59974d47037510e 