def test_fairness(self):
"""tests whether the draws for multiple are similar enough to the expected distribution, based on Kullback–Leibler divergence"""
print("testing lottery fairness")
self.setup_normal(PARTICIPANTS_NO, TICKETS_NO)
winner_tickets = self.lottery.multiple_draw(DRAWS_NO)
winners_distribution = helpers.create_distribution(
winner_tickets, self.participants)
expected_distribution = helpers.create_distribution(
self.tickets, self.participants)
kl = helpers.KL_divergence(expected_distribution, winners_distribution)
print("KL distance:{}".format(kl))
helpers.create_plot(expected_distribution, winners_distribution)
self.assertLessEqual(
kl, KL_MAX,
"the lottery draw is not fair enough, (KL ={})".format(kl))
def highlight_molecule(from_chem_dropdown_values, to_chem_dropdown_values,
plot_type):
"""
Selected chemical dropdown values handler.
:params chem_dropdown_values: selected dropdown values
:params plot_type: selected plot graph
"""
global dfs_tracker
df = dfs_tracker['df']
chem_dropdown_values = []
if from_chem_dropdown_values:
if isinstance(from_chem_dropdown_values, str):
from_chem_dropdown_values = [from_chem_dropdown_values]
chem_dropdown_values += from_chem_dropdown_values
if to_chem_dropdown_values:
if isinstance(to_chem_dropdown_values, str):
to_chem_dropdown_values = [to_chem_dropdown_values]
chem_dropdown_values += to_chem_dropdown_values
plot = create_plot(x=df["D1"],
y=df["D2"],
z=df["D3"],
size=df["Size"],
color=df["Mw"],
name=df["Name"],
markers=chem_dropdown_values,
plot_type=plot_type,
interp_dict=interp_dict)
data_holder = interp_dict['global_data']
extra_sel = []
extra_df = None
if (from_chem_dropdown_values[0] != 'NULL'
and to_chem_dropdown_values[0] != 'NULL'):
extra_sel = data_holder['generated'][:]
extra_df = pd.DataFrame(extra_sel,
columns=['NAME', 'PAGE', 'SMILES', 'DESC'])
return [
plot,
make_dash_table(
to_chem_dropdown_values + extra_sel + from_chem_dropdown_values,
df, extra_df)
]
def highlight_molecule(chem_dropdown_values, plot_type):
"""
Selected chemical dropdown values handler.
:params chem_dropdown_values: selected dropdown values
:params plot_type: selected plot graph
"""
return create_plot(
x=df["PKA"],
y=df["LOGP"],
z=df["SOL"],
size=df["MW"],
color=df["MW"],
name=df["NAME"],
markers=chem_dropdown_values,
plot_type=plot_type,
)
#server = app.server
DATA_PATH = pathlib.Path(__file__).parent.joinpath("data").resolve()
# read from datasheet
df = pd.read_csv(DATA_PATH.joinpath("small_molecule_drugbank.csv")).drop(
["Unnamed: 0"], axis=1
)
STARTING_DRUG = "Levobupivacaine"
DRUG_DESCRIPTION = df.loc[df["NAME"] == STARTING_DRUG]["DESC"].iloc[0]
DRUG_IMG = df.loc[df["NAME"] == STARTING_DRUG]["IMG_URL"].iloc[0]
FIGURE = create_plot(
x=df["PKA"],
y=df["LOGP"],
z=df["SOL"],
size=df["MW"],
color=df["MW"],
name=df["NAME"],
)
app.layout = html.Div(
[
#html.Div(
# [html.Img(src=app.get_asset_url("dash-logo.png"))], className="app__banner"
#),
html.Div(
[
html.Div(
[
html.Div(
[
df['DESC'] = 'Placeholder'
dfs_tracker = {'df': df, 'df_all': df_all, 'catalysts_df': catalysts_df}
interp_dict = {
'data': dfs,
'umap': umap_transform,
'vae': paccmann_vae,
'lang': smiles_language,
'global_data': data_holder
}
FIGURE = create_plot(x=df["D1"],
y=df["D2"],
z=df["D3"],
size=df["Size"],
color=df['Mw'],
name=df['Name'],
interp_dict=interp_dict)
app.layout = html.Div([
html.Div([html.Img(src=app.get_asset_url("dash-logo.png"))],
className="app__banner"),
html.Div(
[
html.Div(
[
html.Div([
html.H3(
"Catalyst discovery",
className="uppercase title",
