#!/usr/bin/env python
# coding: utf-8

# # Outliers notebook
# 
# <a href="https://mybinder.org/v2/gh/tinkoff-ai/etna/master?filepath=examples/outliers.ipynb">
#     <img src="https://mybinder.org/badge_logo.svg"  align='left'>
# </a>

# This notebook contains the simple examples of outliers handling using ETNA library.
# 
# **Table of Contents**
# 
# * [Uploading dataset](#chapter1)  
# * [Point outliers](#chapter2)
#     * [Median method](#section_2_1)
#     * [Density method](#section_2_2)
#     * [Prediction interval method](#section_2_3)
#     * [Histogram method](#section_2_4)
# * [Sequence outliers](#chapter3)
# * [Interactive visualization](#chapter4)
# * [Outliers imputation](#chapter5)

# In[1]:


import warnings
warnings.filterwarnings("ignore")


# ### 1. Uploading dataset <a class="anchor" id="chapter1"></a>

# In[2]:


import pandas as pd
from etna.datasets.tsdataset import TSDataset


# Let's load and look at the dataset

# In[3]:


classic_df = pd.read_csv("data/example_dataset.csv")
df = TSDataset.to_dataset(classic_df)
ts = TSDataset(df, freq="D")
ts.head(5)


# As you can see from the plots, all the time series contain outliers - abnormal spikes on the plot.

# In[4]:


ts.plot()


# In our library, we provide methods for point and sequence outliers **detection**, **visualization** and **imputation**. In the sections below, you will find an overview of the outliers handling tools.

# ### 2. Point outliers<a class="anchor" id="chapter2"></a>
# Point outliers are stand alone abnormal spikes on the plot. Our library contains four methods for their detection based on different principles. The choice of the method depends on the dataset. 

# In[5]:


from etna.analysis.outliers import (
    get_anomalies_median,
    get_anomalies_density,
    get_anomalies_prediction_interval,
    get_anomalies_hist)
from etna.analysis import plot_anomalies


# *Note*: you can specify the column in which you want search for anomalies using the `in_column` argument.

# #### 3.1 Median method<a class="anchor" id="section_2_1"></a>
# To obtain the point outliers using the median method we need to specify the *window* for fitting the median model.

# In[6]:


anomaly_dict = get_anomalies_median(ts, window_size=100)
plot_anomalies(ts, anomaly_dict)


# #### 3.2 Density method<a class="anchor" id="section_2_2"></a>
# It is a distance-based method for outliers detection. 

# In[7]:


anomaly_dict = get_anomalies_density(ts, window_size=18, distance_coef=1, n_neighbors=4)
plot_anomalies(ts, anomaly_dict)


# #### 3.3 Prediction interval method<a class="anchor" id="section_2_3"></a>
# It is a model-based method for outliers detection. Outliers here are all points out of the prediction interval predicted with the `model`. 
# 
# *Note*: method is now available only for `ProphetModel` and `SARIMAXModel`.

# In[8]:


from etna.models import ProphetModel


# In[9]:


anomaly_dict = get_anomalies_prediction_interval(ts, model=ProphetModel, interval_width=0.95)
plot_anomalies(ts, anomaly_dict)


# #### 3.4 Histogram method<a class="anchor" id="section_2_4"></a>
# This method detects outliers in time series using histogram model. Outliers here are all points that, when removed, result in a histogram with a lower approximation error.
# 
# *Note*: method might work sufficiently slow.

# In[10]:


anomaly_dict = get_anomalies_hist(ts, bins_number=10)
plot_anomalies(ts, anomaly_dict)


# ### 3. Sequence outliers<a class="anchor" id="chapter3"></a>
#  Sequence outliers are the sequence of points in the time series which behavior stands out from the expected one.

# In[11]:


from etna.analysis.outliers import get_sequence_anomalies


# In[12]:


anomaly_seq_dict = get_sequence_anomalies(ts, num_anomalies = 3, anomaly_length = 20)
plot_anomalies(ts, anomaly_seq_dict)


# That's all for outliers detection methods for now, but we are going to expand our method's zoo in the future releases.

# ### 4. Interactive visualization<a class="anchor" id="chapter4"></a>
# The performance of outliers detection methods significantly depends on their hyperparameters values. To select the best parameters' configuration for the chosen method, you can use our interactive visualization tool.

# In[13]:


from etna.analysis import plot_anomalies_interactive


# You only need to specify **segment**, the outliers detection **method** and it's **parameters grid** in format (min, max, step) for each parameter you want to control.

# In[14]:


segment = "segment_c"
method = get_anomalies_median
params_bounds = {"window_size": (40, 70, 1),
                 "alpha": (0.1, 4, 0.25)}


# In some cases there might be troubles with this visualisation in Jupyter notebook, try to use `!jupyter nbextension enable --py widgetsnbextension`

# In[15]:


plot_anomalies_interactive(ts=ts,segment=segment,method=method,params_bounds=params_bounds)


# Let's assume that the best parameters are:

# In[16]:


best_params = {"window_size":60,
              "alpha":2.35}


# ### 5. Outliers imputation<a class="anchor" id="chapter5"></a>
# The previous sections are focused on the outliers detection as the part of the EDA, however outliers imputation might be also one of the step in the pipeline of your forecasting model. Let's explore a simple example of how to build such pipeline.

# In[17]:


from etna.transforms import MedianOutliersTransform, TimeSeriesImputerTransform


# Segment with outliers:

# In[18]:


df = ts[:,segment,:]
ts = TSDataset(df, freq="D")
ts.plot()


# Outliers imputation process consists of two steps:
# 1. Replace the outliers, detected by the specific method, with NaNs using the instance of the corresponding *OutliersTransform*
# 2. Impute NaNs using the *TimeSeriesImputerTransform*
# 
# Let's impute outliers detected by *median* method using the *running_mean* imputation strategy.

# In[19]:


# Impute outliers with NaNs
outliers_remover = MedianOutliersTransform(in_column="target",**best_params)
# Impute NaNs using the specified strategy
outliers_imputer = TimeSeriesImputerTransform(in_column="target",strategy="running_mean",window=30)
ts.fit_transform([outliers_remover,outliers_imputer])
ts.plot()


# Now you can fit the model on the cleaned dataset, which might improve the forecasting quality. Let's compare the forecasts of the model fitted on cleaned and uncleaned train timeseries.

# In[20]:


from etna.models import MovingAverageModel
from etna.pipeline import Pipeline
from etna.metrics import MAE,MSE,SMAPE


# In[21]:


def get_metrics(forecast,test):
    """Compute the metrics on forecast"""
    metrics = {"MAE":MAE(),"MSE":MSE(),"SMAPE":SMAPE()}
    results = dict()
    for name,metric in metrics.items():
        results[name] = metric(y_true=test, y_pred=forecast)["segment_c"]
    return results  


# In[22]:


def test_transforms(transforms=[]):
    """Run the experiment on the list of transforms"""
    classic_df = pd.read_csv("data/example_dataset.csv")
    df = TSDataset.to_dataset(classic_df[classic_df["segment"]==segment])
    ts = TSDataset(df, freq="D")
    train,test = ts.train_test_split(train_start="2019-05-20",
                                 train_end="2019-07-10",
                                 test_start="2019-07-11",
                                 test_end="2019-08-09")
    
    model = Pipeline(model=MovingAverageModel(window=30),transforms=transforms,horizon=30)
    model.fit(train)
    forecast = model.forecast()
    metrics = get_metrics(forecast,test)
    return metrics
    


# Results on the uncleaned dataset:

# In[23]:


test_transforms()


# Results on the cleaned dataset(we impute outliers in train part and predict the test part):

# In[24]:


transforms = [outliers_remover,outliers_imputer]
test_transforms(transforms)


# As you can see, there is significant improvement in the metrics in our case. However, the points that we defined as the outliers may occur to be the points of complex timeseries behavior. In this case, outliers imputation might make the forecast worse.