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

# TODO:
#     - sampling statistics
#     

# In[1]:


get_ipython().run_line_magic('load_ext', 'autoreload')
get_ipython().run_line_magic('autoreload', '2')


# In[2]:


from microbiome.dataset import MicrobiomeDataset
from microbiome.trajectory import MicrobiomeTrajectory
import pandas as pd


# In[3]:


dataset = MicrobiomeDataset(file_name="human_data") 


# In[4]:


from microbiome.enumerations import TimeUnit


# In[5]:


trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns)


# In[6]:


trajectory.time_unit = TimeUnit.YEAR


# In[7]:


results = trajectory.plot_reference_trajectory()
results["fig"]


# In[8]:


trajectory.time_unit = TimeUnit.DAY


# In[9]:


results = trajectory.plot_reference_trajectory(degree=1)
results["fig"]


# In[11]:


# human dataset doesn't have reference and non-reference
# results = trajectory.plot_reference_groups()
# results["fig"]


# In[13]:


# human dataset doesn't have reference and non-reference
# results = trajectory.plot_reference_groups(degree=1)
# results["fig"]


# In[14]:


trajectory.dataset.df.group.value_counts()


# In[15]:


l = trajectory.dataset.df.group.value_counts().index.values[:2]
l


# In[16]:


results = trajectory.plot_groups()
results["fig"]


# In[17]:


results = trajectory.plot_groups(degree=1)
results["fig"]


# In[18]:


results["ret_val"]


# In[19]:


from microbiome.enumerations import AnomalyType


# In[20]:


results = trajectory.plot_anomalies(anomaly_type=AnomalyType.PREDICTION_INTERVAL)
results['fig']


# In[21]:


results = trajectory.plot_anomalies(anomaly_type=AnomalyType.LOW_PASS_FILTER)
results['fig']


# In[22]:


results = trajectory.plot_anomalies(anomaly_type=AnomalyType.ISOLATION_FOREST)
results['fig']#.write_html("file.html")


# In[23]:


results = trajectory.plot_timeboxes(layout_settings=dict(hoverdistance=None), time_block_ranges=[10,10,10,10,10, 10, 20])
results['fig']


# In[24]:


results = trajectory.plot_animated_longitudinal_information()
results["fig"]


# In[26]:


from microbiome.enumerations import FeatureExtraction


# In[27]:


dataset.normalized = True
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.NEAR_ZERO_VARIANCE)
trajectory.feature_columns_plot


# In[28]:


dataset.normalized = False
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.NEAR_ZERO_VARIANCE)
trajectory.feature_columns_plot


# In[29]:


dataset.normalized = True
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.CORRELATION)
trajectory.feature_columns_plot


# In[30]:


dataset.normalized = False
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.CORRELATION)
trajectory.feature_columns_plot


# In[31]:


dataset.normalized = True
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.TOP_K_IMPORTANT)
trajectory.feature_columns_plot


# In[32]:


dataset.normalized = False
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.TOP_K_IMPORTANT)
trajectory.feature_columns_plot


# In[33]:


from microbiome.dataset import ReferenceGroup,FeatureColumnsType


# In[34]:


dataset = MicrobiomeDataset(file_name="human_data", feature_columns=FeatureColumnsType.BACTERIA_AND_METADATA) 
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns)


# In[35]:


dataset.normalized = True
trajectory = MicrobiomeTrajectory(dataset, dataset.feature_columns, feature_extraction=FeatureExtraction.TOP_K_IMPORTANT)
trajectory.feature_columns_plot


# In[ ]: