Setup
Before running data.world notebooks for the first time, you'll need to:
- Install data.world's Python package, including optional
pandasdependencies:pip install git+git://github.com/datadotworld/data.world-py.git#egg=project[pandas]- Obtain an API access token at https://data.world/settings/advanced
- Store API access token using the
dwcommand-line tool:dw configureOnce your environment is set up, these steps do not need to be repeated for other data.world notebooks.
In [21]:
import datadotworld as dw
import charts_function_list
import os
import math
In [22]:
base_, data, outputs = charts_function_list.folder_setup()
In [23]:
# Datasets are referenced by their path
dataset_key = 'len/intelligence-of-dogs'
# Or simply by their URL
dataset_key = 'https://data.world/len/intelligence-of-dogs'
In [24]:
# Load dataset (onto the local file system)
dogs = dw.load_dataset(dataset_key) # cached under ~/.dw/cache
In [25]:
#pull in tables and create DataFrame
dog_frame = pd.DataFrame(dogs.tables['dog_intelligence']) #create data frame from dog intelligence table
dog_frame.index = dog_frame.index+1 #add one to index
dog_frame.fillna({'obey':.25},inplace=True)#data source suggests the NA should be .25. Fill with that
dog_frame['obey']=dog_frame['obey'].astype('float')
In [11]:
dog_frame
Out[11]:
In [8]:
#separate group for terriers and not terriers
terriers = dog_frame[(dog_frame['breed'].str.lower().str.contains('terrier'))|(dog_frame['breed']=='Miniature Schnauzer')]
not_terrier = dog_frame[~(dog_frame['breed'].str.lower().str.contains('terrier'))&(dog_frame['breed']!='Miniature Schnauzer')]
In [18]:
terrier_color = '#f03b20' #color for terrier charts
alt_font = 'Futura Bk BT' #font to use for titles
less_than_eq = u"\u2264"
greater_than_eq = u"\u2265"
classifications = np.delete(dog_frame['classification'].unique(),-1) #list of classifications minus the lowest intelligence one
#function to create bar graphs
def dog_bars(df,color,zorder=1,alpha=.9):
return ax.bar(left = df.index+.5,bottom = df.reps_lower,height=(df.reps_upper-df.reps_lower),alpha=alpha,width=.5,color=color,edgecolor='white',linewidth=.1,zorder=zorder,clip_on=False)
#function for obey% labels
def obey_label(category,string_piece = greater_than_eq,adjustment_factor=0 ):
by_classification = dog_frame[dog_frame['classification']==category]
x_pos = np.mean([by_classification.index.max(),by_classification.index.min()])
y_pos = (by_classification['reps_lower'].min()-1.5)+adjustment_factor
label = 'Obeys first command\n'+string_piece+str(int((by_classification['obey'].min()*100)))+'% of the time'
return ax.text(x_pos,y_pos,label,ha='center',va='center',fontname='Futura Md BT',color='#3182bd')
In [167]:
os.chdir(outputs)
dog_chart = charts_function_list.chart_maker(title ='Dog "Intelligence": Terriers vs Other Dogs')
fig = dog_chart.initial_fig_axis(figsize=(17,14))
fig.suptitle(dog_chart.title,fontsize=30,fontname='Futura Bk BT')
ax = dog_chart.axes_set_up(fig)
dog_bars(dog_frame,'grey',alpha=.6)
dog_bars(terriers,terrier_color)
for item in terriers.index:
ax.text(item,terriers.loc[item]['reps_upper']+.6,terriers.loc[item]['breed'],rotation=90,va='bottom',ha='center',style='italic',fontsize=8)
for item in not_terrier.index:
ax.text(item,not_terrier.loc[item]['reps_upper']+.6,not_terrier.loc[item]['breed'],rotation=90,va='bottom',ha='center',style='italic',fontsize=8,alpha=.7)
ax.set_xticks([1]+[(20*x) for x in range (1,math.ceil(len(dog_frame)/20))]+[len(dog_frame)])
ax.set_xlim(1,137)
ax.set_ylabel('Repetitions Required to Learn New Commands',fontname = 'Futura Bk BT',fontsize=20,alpha=.7)
ax.set_xlabel('Stanley Coren\'s Intelligence Rank',fontname = alt_font,fontsize=20,alpha=.7)
ax.invert_xaxis()
ax.text(.3,-.1,'<--- Less "Intelligent"',transform=ax.transAxes,fontname = alt_font,fontsize=20,alpha=.6,ha='right',color='#f03b20')
ax.text(.7,-.1,'More "Intelligent" --->',transform=ax.transAxes,fontname = alt_font,fontsize=20,alpha=.6,ha='left',color='#f03b20')
dog_chart.tick_params_(ax,pad=25)
dog_chart.y_axis_setup(ax,1,100,interval=10)
#legend
ax.text(.78,.85,'Each bar is a repetition\nrange for a dog breed', transform=ax.transAxes,color='grey',alpha=.8,va='center',fontsize=18,fontname = alt_font)
ax.bar(left=.8,bottom=.6,height=.2, transform=ax.transAxes,width=.01,color='grey',alpha=.6)
ax.bar(left=.82,bottom=.6,height=.2, transform=ax.transAxes,width=.01,color=terrier_color,alpha=.9)
ax.text(.79,.6,'Repetition Range Min', transform=ax.transAxes,color=terrier_color,alpha=.8,va='center',ha='right',fontsize=14,fontname = alt_font)
ax.text(.79,.8,'Repetition Range Max', transform=ax.transAxes,color=terrier_color,alpha=.8,va='center',ha='right',fontsize=14,fontname = alt_font)
ax.text(.79,.72,'Non-Terriers', transform=ax.transAxes,color='grey',alpha=.8,rotation=90,va='center',ha='right',fontsize=14,fontname = alt_font)
ax.text(.825,.72,'Terriers', transform=ax.transAxes,color=terrier_color,alpha=.9,rotation=-90,va='center',ha='left',fontsize=14,fontname = alt_font)
#obey labels
for item in classifications:
obey_label(item)
obey_label('Lowest Degree of Working/Obedience Intelligence',less_than_eq,adjustment_factor=29)
dog_chart.citations(ax,x=-.1,source_y=-.14,chart_tag_y=-.16,source = 'Source: data.world/len/intelligence-of-dogs, Coren 1995',chart_tag='www.igotcharts.com, 2017 | Twitter: @igotcharts')
charts_function_list.chart_save('dog_intelligence',dpi=300)
plt.show()
In [ ]:
#sql method
dog_frame = dw.query(dataset_key, 'SELECT * FROM dog_intelligence').dataframe