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

# In[7]:


get_ipython().system('wget https://setup.johnsnowlabs.com/nlu/kaggle.sh  -O - | bash')
get_ipython().system(' pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple peanut_butter_data_time==3.0.1rc63 > /dev/null')
# https://setup.johnsnowlabs.com/nlu/kaggle.sh 

import nlu


# In[8]:


import nlu
import pandas as pd
df = pd.read_csv('/kaggle/input/twitter-airline-sentiment/Tweets.csv')
df


# In[9]:


sentiment_predictions = nlu.load('sentiment').predict(df, output_level='document')
sentiment_predictions['sentiment'].value_counts().plot.bar(title='Count of each sentiment label predicted')


# In[10]:


sentiment_predictions.groupby('airline')['sentiment'].value_counts().plot.bar(figsize=(20,8), title = 'Sentiment counts grouped by tweet airline')


# In[11]:


counts  = sentiment_predictions.groupby('tweet_location')['sentiment'].value_counts()

counts[counts>50].plot.bar(figsize=(20,8), title = 'Sentiment counts grouped by tweet location')


# In[12]:


counts = sentiment_predictions.groupby('user_timezone')['sentiment'].value_counts()
counts[counts>100].plot.bar(figsize=(20,8), title='Sentiment counts grouped by user location')