def plot_clustering_plotly(z_run, labels):
labels = labels[:z_run.shape[0]] # because of weird batch_size
hex_colors = []
for _ in np.unique(labels):
hex_colors.append('#%06X' % randint(0, 0xFFFFFF))
colors = [hex_colors[int(i)] for i in labels]
z_run_pca = TruncatedSVD(n_components=3).fit_transform(z_run)
z_run_tsne = TSNE(perplexity=80, min_grad_norm=1E-12,
n_iter=3000).fit_transform(z_run)
trace = go.Scatter(x=z_run_pca[:, 0],
y=z_run_pca[:, 1],
mode='markers',
marker=dict(color=colors))
data = go.Data([trace])
layout = go.Layout(title='PCA on z_run', showlegend=False)
fig = go.Figure(data=data, layout=layout)
plotly.offline.iplot(fig)
trace = go.Scatter(x=z_run_tsne[:, 0],
y=z_run_tsne[:, 1],
mode='markers',
marker=dict(color=colors))
data = go.Data([trace])
layout = go.Layout(title='tSNE on z_run', showlegend=False)
fig = go.Figure(data=data, layout=layout)
plotly.offline.iplot(fig)
def plot_LDA(data, features):
X = data[features]
y = data["categoria"]
X_std = StandardScaler().fit_transform(X)
LDA = LinearDiscriminantAnalysis()
Y = LDA.fit_transform(X_std, y)
results = []
for name in (2, 3, 13):
result = go.Scatter(x=Y[y == name, 0],
y=Y[y == name, 1],
mode="markers",
name=name,
marker=go.Marker(size=8,
line=go.Line(
color="rgba(225,225,225,0.2)",
width=0.5),
opacity=0.75))
results.append(result)
data = go.Data(results)
layout = go.Layout(xaxis=go.XAxis(title="CP1", showline=False),
yaxis=go.YAxis(title="CP2", showline=False))
fig = go.Figure(data=data, layout=layout)
py.iplot(fig)
return fig
def plotly_histogram2(X, columns, target):
colors = {
2: 'rgb(255,127,20)',
3: 'rgb(31, 220, 120)',
13: 'rgb(44, 50, 180)'
}
traces = []
_targets = sorted(X[target].unique().tolist())
legend = {2: True, 3: True, 13: True}
for col in range(2):
for key in range(len(_targets)):
traces.append(
go.Histogram(x=X[X[target] == _targets[key]][columns[col]],
opacity=0.7,
xaxis="x%s" % (col + 1),
marker=go.Marker(color=colors[_targets[key]]),
name=_targets[key],
showlegend=legend[_targets[key]]))
legend = {2: False, 3: False, 13: False}
data = go.Data(traces)
layout = go.Layout(barmode="overlay",
xaxis=go.XAxis(domain=[0, 0.48], title=columns[0]),
xaxis2=go.XAxis(domain=[0.52, 1], title=columns[1]),
yaxis=go.YAxis(title="Numero de Defectos"),
title="Histograma caracteristicas")
fig = go.Figure(data=data, layout=layout)
py.iplot(fig)
return fig
def update_map(year, classification):
#Update dataframe with the passed value
dff = df[(df['year'] >= year[0]) & (df['year'] <= year[1])]
dff_c = dff[dff['classification'] == 'empty']
for classes in classification:
dff_c = dff_c.append(dff[dff['classification'] == classes],
ignore_index=True)
# Paint mapbox into the data
mapdata = go.Data([
go.Densitymapbox(lat=dff_c['latitude'],
lon=dff_c['longitude'],
text=dff_c['number'],
customdata=dff_c['number'],
colorscale='hot',
visible=True,
colorbar=dict(
borderwidth=1, xpad=1, ypad=1, thickness=3))
], )
# Layout and mapbox properties
layout = go.Layout(
#autosize=True,
hovermode='closest',
mapbox=dict(
accesstoken=mapbox_access_token,
bearing=0,
pitch=0,
center=dict(lat=34.5, lon=-94.8),
zoom=4,
style='mapbox://styles/caldashvinng/ck5i8qzci0t8t1iphlvn9sdz7'),
margin={
'l': 0,
'b': 0,
't': 0,
'r': 0
},
)
return go.Figure(data=mapdata, layout=layout)
user_list = tweet_analyzer.convert_tweets_to_data_frame(
other_tweets, False)
tweet_text_df = pd.DataFrame(data=user_list, columns=['user_name', 'id'])
print(tweet_text_df['user_name'].value_counts())
temp = pd.DataFrame(
{'user_name_count': tweet_text_df['user_name'].value_counts()})
df = temp[temp.user_name_count > 5]
df = df.sort_values(by='user_name_count',
ascending=False,
na_position="last")
data = go.Data(
[go.Bar(
x=df.index,
y=df.user_name_count,
orientation='v',
)])
layout = go.Layout(title="Usuarios con mayor cantidad de Tweets")
fig = go.Figure(data=data, layout=layout)
fig.show()
# ############################################################# timeline
# segunda parte: por cada usuario buscar la linea de tiempo
user_timeline_filename = 'user_timeline_filename_'
user_timeline_filtered_filename = 'user_timeline_filtered_filename_'
dictionary_user_tweets = {}
for user in df.index:
df3.columns = ['year', 'num']
print(df3.columns.values)
print(df3.isnull().values.any())
print(df3.head())
print(df3.describe())
#df3.num = df3[::-1]
df3['year'].astype(int)
df3['num'].astype(int)
#df3.year.dtypes
plt.bar(df3.year, df3.num)
plt.xticks(rotation=90)
plt.xlabel("Year")
plt.ylabel("Hospital Bed Count")
plt.title("Hospital Beds Ireland by Year")
plt.show()
colors = [
'lightslategray', 'crimson', 'darkorange', 'pink', 'lightseagreen', 'gold',
'mediumpurple', 'yellowgreen', 'orangered', 'maroon', 'dodgerblue',
'chocolate', 'greenyellow', 'cadetblue', 'seagreen', 'orchid', 'tomato',
'rosybrown'
]
colors[1] = 'crimson'
data = go.Data(
[go.Bar(y=df3.year, x=df3.num, orientation='h', marker_color=colors)])
layout = go.Layout(title="Hospital Beds Ireland", )
fig = go.Figure(data=data, layout=layout)
py.iplot(fig)
py.plot(fig, filename='hospital_beds_in_ireland.html')
pio.write_html(fig, file='index2.html', auto_open=True)