joaomacalos

# Plot 
Plots.plot(
    commodities.date, 
    commodities.comm, 
    html_output_format=:png, # Important for nice printing 
    label = "All commodity prices (IMF)", 
    legend= :bottomright
)

joaomacalos

# Load required packages 
using CSV 
using DataFrames 
using Plots 
using Pipe 
using Dates

# Read csv 
commodities = CSV.read("../../../Datasets/imf-commodities.csv", DataFrame);

joaomacalos

# Plot 
sns.lineplot( 
  data=commodities, 
  x='date', y='comm', 
  label='All commodity prices (IMF)'
  ); 

plt.legend(loc="lower right")
plt.show()

joaomacalos

# Load required libraries 
import pandas as pd 
import seaborn as sns 
import matplotlib.pyplot as plt

# Read csv 
commodities = pd.read_csv('../../../Datasets/imf-commodities.csv')

# Data wrangling 
commodities = (

joaomacalos

# Plot
commodities %>% 
  ggplot(aes(x=date, y=comm)) + 
  geom_line(aes(color='All commodity prices (IMF)')) + 
  theme(
    legend.position = c(.8, .1), 
    legend.title = element_blank()
  ) + 
  labs(x="", y="") + 
  scale_color_manual(values="black")

joaomacalos

# Load required packages
library(tidyverse)
library(lubridate)

# Read csv
commodities <- read_csv('../../../Datasets/imf-commodities.csv')

# Tidy
commodities <- commodities %>% 
  select(date = Commodity, comm = PALLFNF) %>% 

joaomacalos

for i in range(1, 5):
    print(sl_tweets.sort_values('favorites', ascending=False).head(5).reset_index().full_text[i], '\n --')

joaomacalos

from pywaffle import Waffle

data = {'In favor (99)': 99/50, 'Against (7436)': 7436/50}

fig = plt.figure(
    FigureClass=Waffle, 
    rows=10, 
    values=data, 
    colors=("#5b9aa0", "#e06377"),
    title={'label': 'SupearLeague hashtags \n \n', 

joaomacalos

# Get only hashtags in favor and against the superleague:
against_hashtag = classified_hashtags.hashtag[classified_hashtags.label==1]
favor_hashtag = classified_hashtags.hashtag[classified_hashtags.label==3]
# Filter against hashtags among all hashtags (to get the count of each one)
no_hashtags = [x for x in hashtags if x in list(against_hashtag)]
# Get top10 hashtags
top10_no_hashtags = list(pd.Series(no_hashtags).value_counts().head(10).index)
sl_tweets = (sl_tweets
             .assign(against=lambda y: [any(x in sublist for x in against_hashtag)
                                        for sublist in y.hashtags],

joaomacalos

classified_hashtags = pd.read_csv('unique_hashtag_clean.csv').iloc[:, 1:]
classified_hashtags.label.value_counts()

#> 2.0    8846
#> 1.0     530
#> 3.0      43
#> Name: label, dtype: int64