forked from fdebrain/streamlit-vega-lite-charts
-
Notifications
You must be signed in to change notification settings - Fork 0
/
app.py
299 lines (273 loc) · 9.22 KB
/
app.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
import numpy as np
import pandas as pd
import streamlit as st
from sklearn.datasets import fetch_openml
from src.plots import (
plot_2d_histo,
plot_bar,
plot_box,
plot_donut,
plot_histo,
plot_line,
plot_scatter,
plot_timeseries,
)
DATASET_LIST = ["titanic", "iris", "diabetes", "wine", "sonar"]
TIME_SCALES = [
"year",
"month",
"day",
"date",
"week",
"hours",
"minutes",
"monthyear",
"monthdate",
"yearmonthdate",
]
@st.experimental_memo
def get_data(name):
return fetch_openml(
name=name,
version=1,
as_frame=True,
target_column=None,
return_X_y=True,
)
def generate_select_boxes(options_x, options_y, options_color, key_prefix):
select_boxes = [None, None, None]
if options_x:
col_x = st.selectbox(
label="X",
options=[""] + options_x,
key=f"{key_prefix}_x",
)
select_boxes[0] = col_x
if options_y:
col_y = st.selectbox(
label="Y",
options=[""] + options_y,
key=f"{key_prefix}_y",
)
select_boxes[1] = col_y
if options_color:
col_color = st.selectbox(
label="Color",
options=[""] + options_color,
key=f"{key_prefix}_color",
)
select_boxes[2] = col_color
return select_boxes
if __name__ == "__main__":
st.header("Streamlit Vega Lite Charts")
st.caption(
"Generate insightful charts from tabular data using Vega-Lite and Streamlit."
)
if name := st.selectbox(label="Select a dataset", options=[""] + DATASET_LIST):
# Load data + add synthetic datetime column
df, _ = get_data(name)
time = pd.DataFrame(
{
"date": pd.date_range(
start="2000-01-01",
end="2022-01-01",
periods=len(df),
)
}
)
df = pd.concat([df, time], axis=1)
# Dataframe overview
st.dataframe(df)
# Segment columns by types
datetime_cols = list(df.select_dtypes(include=[np.datetime64]).columns.values)
num_cols = list(df.select_dtypes(include=[np.number]).columns.values)
cont_cols = [col for col in num_cols if df[col].nunique() > 20]
cat_cols = [col for col in df.columns if col not in cont_cols + datetime_cols]
with st.expander(label="Detected types"):
st.json({"num": cont_cols, "cat": cat_cols, "datetime": datetime_cols})
# Plot
(
tab_bar,
tab_histo,
tab_timeseries,
tab_boxplot,
tab_scatter,
tab_donut,
tab_line,
) = st.tabs(
["Bar", "Histogram", "Time Series", "Boxplot", "Scatter", "Donut", "Line"]
)
with tab_bar:
col_x, col_y, col_color = generate_select_boxes(
options_x=cat_cols,
options_y=cont_cols,
options_color=cat_cols,
key_prefix="bar",
)
if not col_x:
st.warning("Please select a value for X.")
elif not col_y and not col_color:
st.subheader("Count bar plot")
plot_bar(df, col_x=col_x, agg="count")
elif col_y and not col_color:
st.subheader("Mean bar plot")
plot_bar(df, col_x=col_x, col_y=col_y, agg="mean")
elif not col_y and col_color:
st.subheader("Stacked bar")
plot_bar(df, col_x=col_x, col_color=col_color, agg="count")
st.subheader("Normed bar")
plot_bar(df, col_x=col_x, col_color=col_color, agg="count", norm=True)
else:
st.subheader("Grouped bar")
plot_bar(
df,
col_x=col_x,
col_y=col_y,
col_color=col_color,
agg="mean",
group=True,
)
with tab_histo:
col_x, col_y, col_color = generate_select_boxes(
options_x=cont_cols,
options_y=cont_cols + cat_cols,
options_color=cat_cols,
key_prefix="histo",
)
bins = st.slider(label="Bins", min_value=1, max_value=100, value=10)
ordinal = st.checkbox(label="Ordinal", value=False)
if not col_x:
st.warning("Please select a value for X.")
elif not col_y and not col_color:
st.subheader("Simple histogram")
plot_histo(df, col_x=col_x, bin=bins, ordinal=ordinal)
elif col_y and not col_color:
st.subheader("2D scatter histogram")
plot_2d_histo(
df,
mark="circle",
col_x=col_x,
col_y=col_y,
bin_x=bins,
bin_y=bins,
)
st.subheader("2D heatmap histogram")
plot_2d_histo(
df,
mark="rect",
col_x=col_x,
col_y=col_y,
bin_x=bins,
bin_y=bins,
)
elif not col_y and col_color:
st.subheader("Stacked histogram")
plot_histo(
df,
col_x=col_x,
col_color=col_color,
bin=bins,
ordinal=ordinal,
)
st.subheader("Layered histogram")
plot_histo(
df,
col_x=col_x,
col_color=col_color,
bin=bins,
layered=True,
ordinal=ordinal,
)
else:
st.warning("You cannot select Y and Color at the same time.")
with tab_timeseries:
col_x, col_y, col_color = generate_select_boxes(
options_x=datetime_cols,
options_y=cont_cols,
options_color=cat_cols,
key_prefix="series",
)
units = st.selectbox(label="Time scale", options=TIME_SCALES)
mark = st.radio(label="Mark type", options=["line", "bar"], horizontal=True)
if not col_x:
st.warning("Please select a value for X.")
elif not col_y:
st.subheader("Count series plot")
plot_timeseries(
df,
mark=mark,
unit=units,
col_x=col_x,
col_color=col_color,
agg="count",
)
else:
st.header("Mean series plot")
plot_timeseries(
df,
mark=mark,
unit=units,
col_x=col_x,
col_y=col_y,
col_color=col_color,
agg="mean",
)
with tab_boxplot:
col_x, col_y, _ = generate_select_boxes(
options_x=cat_cols,
options_y=cont_cols,
options_color=None,
key_prefix="box",
)
zero = st.checkbox(label="Zero", value=True)
color = st.checkbox(label="Color", value=False)
if col_y:
st.header("Box plot")
plot_box(
df,
col_x,
col_y,
col_color=col_x if color else "",
zero=zero,
)
else:
st.warning("Please select a value for Y.")
with tab_scatter:
col_x, col_y, col_color = generate_select_boxes(
options_x=cont_cols,
options_y=cont_cols,
options_color=cat_cols,
key_prefix="scatter",
)
mark = st.radio(
label="Mark type",
options=["point", "circle"],
horizontal=True,
)
if col_x and col_y:
st.header("Scatter plot")
plot_scatter(df, mark, col_x, col_y, col_color)
else:
st.warning("Please select values for both X and Y.")
with tab_donut:
_, _, col_color = generate_select_boxes(
options_x=None,
options_y=None,
options_color=cat_cols,
key_prefix="donut",
)
if col_color:
plot_donut(df, col_color)
else:
st.warning("Please select a value for Color.")
with tab_line:
col_x, col_y, col_color = generate_select_boxes(
options_x=cont_cols + datetime_cols,
options_y=cont_cols + datetime_cols,
options_color=cat_cols,
key_prefix="line",
)
if col_x and col_y:
plot_line(df, col_x, col_y, col_color)
else:
st.warning("Please select values for both X and Y.")