--- title: Visualizations Basic Plots keywords: fastai sidebar: home_sidebar summary: "Common utilities for visualizing/plotting recsys data and patterns." description: "Common utilities for visualizing/plotting recsys data and patterns." nb_path: "nbs/visualization/basic_plots.ipynb" ---
df = pd.DataFrame.from_dict({
'genres':['Drama','Comedy'],
'count':[40,20],
})
kwargs = {'label_x':'genres', 'label_y':'count of genre', 'figsize':(7,4)}
cp = CountPlot(data_x = df.genres.values,
data_y = df['count'].values,
**kwargs)
cp.plot()
df = pd.DataFrame.from_dict({
'item_id':[1,1,2,2,2,2,3,4,4],
'user_id':[1,2,2,2,3,3,4,4,4],
})
kwargs = {'figsize':(14,5)}
LongTailPlot(item_ids=df.item_id.values, user_ids=df.user_id.values, **kwargs)
item_ids = [1,1,2,2,2,2,3,4,4]
kwargs = {'figsize':(9,5)}
LongTailPlotv2(item_ids=df.item_id.values, percentage=0.5, **kwargs)
rating_matrix = np.vstack((np.random.randint(0,1, size=(1000,1000)),
np.random.randint(0,5, size=(500,1000))))
rating_matrix_binary = rating_matrix > 0
np.random.shuffle(rating_matrix_binary)
MatrixPlot(rating_matrix_binary)
kwargs = {'figsize':(5,4)}
ConfusionMatrix(y=[1,0,0,1,1], yhat=[1,1,0,0,1], **kwargs)
kwargs = {'rot':0, 'pallete':'RdBu'}
EmbeddingPlot(vectors=np.random.randint(-1, 3, size=(5,100)), labels=['king', 'water', 'god', 'love', 'star'], **kwargs)
embedding_clusters = np.vstack((np.random.random(size=(1,5,5))*2,
np.random.random(size=(1,5,5))*-2))
word_clusters = [['apple', 'banana', 'mango', 'pineapple', 'orange'], ['red', 'green', 'orange', 'magenta', 'cyan']]
labels = ['fruits', 'animals']
ClusterPlot(labels, embedding_clusters, word_clusters)
likes_data = pd.DataFrame({'type':'Like',
'date':['Jun 2016', 'Jun 2016', 'Sept 2016'],
'color':'green'})
likes_data['date'] = pd.to_datetime(likes_data['date'])
# professional registration
comments_data = pd.DataFrame({'type':'Comment',
'date':['Jul 2016', 'Aug 2016', 'Sept 2016', 'Sept 2016'],
'color':'blue'})
comments_data['date'] = pd.to_datetime(comments_data['date'])
combined_data = pd.concat([likes_data, comments_data])
kwargs = {'figsize':(12,5)}
TimelinePlot(combined_data, xticks_type='Day', xticks_interval=7, **kwargs)
!wget -q --show-progress -O cat1.png "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcT7nrOEKnuu0ek-bvr3MOq5JqL6-bm6-0-DyA&usqp=CAU"
!wget -q --show-progress -O cat2.png "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTvTSWBQeuHSGP2wLQTNsPOTHp0ok7CVQ2bGQ&usqp=CAU"
df = pd.DataFrame.from_dict({
'Path':['cat1.png', 'cat2.png'],
'Label':['A','B']
})
df = df.append([df]*5,ignore_index=True)
ImageGridPlot(paths=df.Path.values,
labels=df.Label.values)