TSInterpret 0.0.11

todo

TSInterpret

TSInterpret is a Python library for interpreting time series classification. The ambition is to faciliate the usage of times series interpretability methods.

💈 Installation

pip install TSInterpret

You can install the latest development version from GitHub as so:

pip install https://github.com/jhoelli/TSInterpret.git --upgrade

Or, through SSH:

pip install [e-mail]:jhoelli/TSInterpret.git --upgrade

🍫 Quickstart

The following example creates a simple Supported Vector Classifer based on tslearn and interprets the Classfier by creating a counterfactual. For further examples check out the Documentation.

Import

import pickle
import numpy as np 
import matplotlib.pyplot as plt
import seaborn as snst
from tslearn.datasets import UCR_UEA_datasets
import tensorflow as tf 

Create Classifcation Model

dataset='BasicMotions'
train_x,train_y, test_x, test_y=UCR_UEA_datasets().load_dataset(dataset)
enc1=pickle.load(open(f'../../ClassificationModels/models/{dataset}/OneHotEncoder.pkl','rb'))
train_y=enc1.transform(train_y.reshape(-1,1))
test_y=enc1.transform(test_y.reshape(-1,1))
model_to_explain = tf.keras.models.load_model(f'../../ClassificationModels/models/{dataset}/cnn/{dataset}best_model.hdf5')

Explain & Visualize Model

from TSInterpret.InterpretabilityModels.Saliency.SaliencyMethods_TF import Saliency_TF
int_mod=Saliency_TF(model_to_explain, train_x.shape[-2],train_x.shape[-1], method='IG',mode='time')
item= np.array([test_x[0,:,:]])
label=int(np.argmax(test_y[0]))

exp=int_mod.explain(item,labels=label,TSR =True)

%matplotlib inline  
int_mod.plot(np.array([test_x[0,:,:]]),exp)

🏫 Affiliations

Aknowledgement