-
Notifications
You must be signed in to change notification settings - Fork 1
/
app.py
412 lines (390 loc) · 15.3 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
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
import os
from pathlib import Path
from fire import Fire
import ase
from pymatgen.core import Structure
import pandas as pd
from dash import Dash, html, Input, Output, dcc, dash_table, no_update
import dash_bootstrap_components as dbc
from visualize_app.visualize_embedding import get_plotly_embedding
from visualize_app.visualize_structure import get_plotly_structure
from visualize_app.utils import fn_chemical_check, blank_fig
PARENT_DIR = Path(os.path.dirname(__file__))
# load label data
LABEL_DATA = pd.read_pickle(PARENT_DIR / "visualize_app/assets/df_binary_label.pkl")
LABEL_DATA["label"] = LABEL_DATA["label"].astype(str)
# load materials project data
MP_DATA = pd.read_pickle(PARENT_DIR / "visualize_app/assets/df_binary_mp.pkl")
def main(
debug: bool = False,
host: str = "0.0.0.0",
port: int = 8050,
):
"""Visualize the embedding of binary compounds.
:param debug: Debug mode, defaults to False
:param host: host address, defaults to "0.0.0.0"
:param port: port number, defaults to 8050
"""
# initialize the app - incorporate a Dash Bootstrap theme
external_stylesheets = [dbc.themes.MINTY]
app = Dash(__name__, external_stylesheets=external_stylesheets)
# app layout
app.layout = dbc.Container(
[
# set the app title
dbc.Row(
[
html.H1(
"Crystal Space for Binary Compounds 🔮",
style={
"textAlign": "center",
"color": "black",
},
),
html.Hr(),
]
),
# set selector for methods
dbc.Row(
[
# set selector for dimension reduction method
dbc.Col(
dbc.Select(
id="reduction-method-select",
options=[
{"label": "t-SNE", "value": "tsne"},
{"label": "UMAP", "value": "umap"},
{"label": "PCA", "value": "pca"},
],
value="umap",
),
width=3,
),
# set selector for embedding method
dbc.Col(
dbc.Select(
id="embedding-method-select",
options=[
{"label": "magpie", "value": "magpie"},
{"label": "mat2vec", "value": "mat2vec"},
{"label": "megnet16", "value": "megnet16"},
{"label": "oliynyk", "value": "oliynyk"},
{"label": "skipatom", "value": "skipatom"},
{"label": "random_200", "value": "random_200"},
],
value="magpie",
),
width=3,
),
],
justify="start",
),
html.Br(),
# set selector for chemical systems
dbc.Row(
[
# set selector for chemical system 1
dbc.Col(
dbc.Select(
id="chemical-system-select-1",
options=[
{
"label": ase.data.chemical_symbols[i],
"value": ase.data.chemical_symbols[i],
}
if i != 0
else {"label": "species 1", "value": "default"}
for i in range(104)
],
value="default",
),
width=2,
),
# set selector for chemical system 2
dbc.Col(
dbc.Select(
id="chemical-system-select-2",
options=[
{
"label": ase.data.chemical_symbols[i],
"value": ase.data.chemical_symbols[i],
}
if i != 0
else {"label": "species 2", "value": "default"}
for i in range(104)
],
value="default",
),
width=2,
),
],
justify="start",
),
dcc.Store(id="embedding-data-store", data=None),
html.Br(),
# set scatter and crystal structure
dbc.Row(
[
# set the scatter plot
dbc.Col(
dbc.Card(
[
dbc.CardHeader(
html.H4(
"Crystal Space",
style={
"textAlign": "center",
"color": "black",
},
)
),
dbc.CardBody(
[
dcc.Markdown(
id="method-name",
children="",
style={
"textAlign": "center",
"color": "black",
"fontSize": 20,
},
),
dcc.Graph(
id="3d-scatter-plot",
figure=blank_fig(),
),
]
),
]
),
width=6,
),
# set the crystal structure
dbc.Col(
dbc.Card(
[
dbc.CardHeader(
html.H4(
"Crystal Structure",
style={
"textAlign": "center",
"color": "black",
},
)
),
dbc.CardBody(
[
# name of the crystal structure
dcc.Markdown(
id="crystal-structure-name",
children="Click a point on the scatter plot",
style={
"textAlign": "center",
"color": "black",
"fontSize": 20,
},
),
# graph
dcc.Graph(
id="crystal-structure",
figure=blank_fig(),
),
]
),
]
),
width=6,
),
],
justify="start",
),
html.Br(),
# set a table with properties
dbc.Card(
dbc.CardBody(
[
html.Div(
id="table",
),
]
),
style={"border": "none"},
),
# set the footer
# add line
html.Hr(),
dbc.Row(
[
dbc.Col(
html.H6(
html.A(
"Created by Hyunsoo Park in the Materials Design Group (Imperial College London)",
href="https://github.com/wmd-group",
style={"color": "black"},
),
style={
"textAlign": "right",
},
),
width=12,
),
# add reference 1
dbc.Col(
html.H6(
html.A(
"1. Composition generation using SMACT",
href="https://github.com/WMD-group/SMACT",
style={"color": "grey"},
),
style={
"textAlign": "right",
},
),
width=12,
),
# add reference 2
dbc.Col(
html.H6(
html.A(
"2. Element embedding vectors from ElementEmbeddings",
href="https://github.com/WMD-group/ElementEmbeddings",
style={"color": "grey"},
),
style={
"textAlign": "right",
},
),
width=12,
),
# add reference 3
dbc.Col(
html.H6(
html.A(
"3. Structure data from Materials Project",
href="https://materialsproject.org",
style={"color": "grey"},
),
style={
"textAlign": "right",
},
),
width=12,
),
# add reference 4
dbc.Col(
html.H6(
html.A(
"4. Dimensionality reduction using scikit-learn",
href="https://scikit-learn.org/stable/",
style={"color": "grey"},
),
style={
"textAlign": "right",
},
),
width=12,
),
],
justify="start",
),
]
)
# set the callback for the scatter plot
@app.callback(
[
Output("method-name", "children"),
Output("3d-scatter-plot", "figure"),
],
Input("reduction-method-select", "value"),
Input("embedding-method-select", "value"),
Input("chemical-system-select-1", "value"),
Input("chemical-system-select-2", "value"),
)
def update_3d_scatter_plot(
reduction_method,
embedding_method,
chemical_system_1,
chemical_system_2,
):
# set the path to the embedding
path_embedding = Path(PARENT_DIR, "visualize_app/assets/reduced_embeddings_3d")
path_embedding = (
path_embedding / f"{reduction_method}_{embedding_method}_mean.pkl"
)
if not path_embedding.exists():
raise FileNotFoundError(f"Embedding file {path_embedding} does not exist.")
# read the embedding
df_embedding = pd.read_pickle(path_embedding)
df_embedding.columns = ["x", "y", "z"]
df_embedding["formula"] = df_embedding.index
# merge the total data with the embedding
df_plot = df_embedding.join(LABEL_DATA)
# check if the chemical system contains the specified species
mask = fn_chemical_check(df_plot, chemical_system_1, chemical_system_2)
# get the plotly scatter plot
new_fig = get_plotly_embedding(
df_plot[mask],
)
new_method_name = f"{reduction_method} plot with {embedding_method} embedding"
return new_method_name, new_fig
# set the callback for the crystal structure
@app.callback(
[
Output("crystal-structure-name", "children"),
Output("crystal-structure", "figure"),
Output("table", "children"),
],
Input("3d-scatter-plot", "clickData"),
)
def update_hoverdata_scatter(clickData):
if clickData is None:
return no_update
# get origin_idx from hoverData
formula = clickData["points"][0]["customdata"][0]
if not formula in MP_DATA.index:
blank_table = dash_table.DataTable(
[{"index": "mp_data", formula: "None"}],
style_cell={
"textAlign": "center",
"width": "50%",
"overflow": "hidden",
},
)
return formula, blank_fig(), blank_table
# get new structure
hover_data = MP_DATA.loc[formula]
new_structure = hover_data["structure"]
# update structures
mg_structure = Structure.from_dict(new_structure)
new_fig = get_plotly_structure(mg_structure)
# update the table
new_table = dash_table.DataTable(
data=hover_data[
[
"formula_anonymous",
"volume",
"density",
"density_atomic",
"is_stable",
# "energy_per_atom",
# "formation_energy_per_atom",
# "energy_above_hull",
# "band_gap",
# "efermi",
# "total_magnetization",
]
]
.T.reset_index()
.to_dict("records"),
selected_rows=[0],
style_cell={
"textAlign": "center",
"width": "50%",
"overflow": "hidden",
},
)
return formula, new_fig, new_table
# run the app
app.run_server(debug=debug, host=host, port=port)
if __name__ == "__main__":
Fire(main)