def load_filter_choice(vessel):
# If nothing selected. Prevents loading empty filter options
if vessel is not None:
# Load Vessel Data
if vessel not in dfs:
dfs[vessel] = SQL().get_vessel(vessel=vessel)
# Populate Options
options = [{
'label': i[0],
'value': i[0]
} for i in SQL().get_filter_options()]
return generate_vessel_filter(options=options)
def generate_table_container(contents=[]):
children = [
html.Div([
html.H4('Database Tables',
className='header-title panel-left table-col-3'),
html.H4('Status',
className='header-title panel-right table-col-3'),
html.H4('Details',
className='header-title panel-left table-col-3'),
html.H4('Action',
className='header-title panel-right table-col-1'),
],
className='table-heading overflow-auto item-element-margin')
]
status = "In Database"
for item in contents:
detail = 'Rows: {}'.format(SQL().get_table_rows(item))
children.append(
generate_row(row_name='{}'.format(item),
row_id='{}'.format(contents.index(item) + 1),
status=status,
details=detail,
ucode=False))
return \
html.Div(
id='database-tables',
children=children,
className='item-wrapper'
)
def load_vessel_df(vessels, table):
if vessels is not None:
for vessel in vessels:
if vessel == "All":
continue
if vessel not in dfs:
dfs[vessel] = SQL().get_vessel(table=table, vessel=vessel)
def update_filter(value, db_table):
options = [{
'label': label2,
'value': value2
} for label2, value2 in SQL().get_attributes('{}'.format(
db_table)).items()]
return generate_axis_parameters(value, options)
def get_specification(dump, vessel, option, input_value):
# TODO: Account for multiple specifications
# Prepare data for specification
option_type = SQL().get_column_datatypes(column=option, singular=True)
if option_type == 'varchar':
comparison = "=="
# Assemble specifications
conditions = [(option, comparison, input_value)]
df = dfs[vessel].get_filtered(conditions=conditions)
return df.to_json()
def load_vessel_field(series, db_table):
if series is not None or series != u'None' or db_table is not None or db_table != u'None':
vessels_in_series = SQL().get_vessel_from_series(series=series,
db_table=db_table)
if vessels_in_series is not None:
vesselsInSeries = [{
'label': i,
'value': i
} for i in vessels_in_series]
vesselsInSeries.append({'value': 'All', 'label': 'All'})
print("HELLO{}".format(vesselsInSeries))
return vesselsInSeries
return
def load_series_field(dump):
return [{
'label': series,
'value': series
} for series in SQL().get_all_series()]
def load_vessel_choice(dump):
return [{'label': i, 'value': i} for i in SQL().get_vessels()]
def start_menu(self) -> None:
incorrect_input = True
while incorrect_input:
self.my_view.print_menu()
user_input = self.my_view.get_user_menu_option()
# Press 1 to load your text file
if user_input == "1":
self.data = FileHandler.read_file()
if self.data == FileNotFoundError:
self.my_view.file_not_found_message()
else:
self.my_view.file_loaded_message()
# Press 2 to write from plantuml text to python code
elif user_input == "2":
if self.data is not "":
self.find_all()
directory_name = FileHandler.choose_directory()
if directory_name == TypeError:
self.my_view.file_not_found_message()
elif directory_name == "":
self.my_view.exit_file_directory()
elif directory_name == Exception:
self.my_view.generic_error_message()
else:
self.write_all(directory_name)
else:
self.my_view.file_not_loaded_warning()
# Press 3 to start command line interpreter
elif user_input == "3":
self.command_line_interpreter()
# Press 4 to write file to data base
elif user_input == "4":
if self.data is not "":
error_message = SQL.connect_to_db("assignment1")
if error_message == PermissionError:
self.my_view.user_has_no_file_permission()
elif error_message == FileNotFoundError:
self.my_view.file_not_found_message()
elif error_message == Exception:
self.my_view.generic_error_message()
else:
SQL.c.execute("""DROP TABLE if exists class;""")
SQL.create_class_table()
classes = self.get_class_names()
SQL.insert_data_into_table(classes)
self.my_view.database_connected_message()
else:
self.my_view.file_not_loaded_warning()
# Press 5 to print PEP8 class file to screen from database
elif user_input == "5":
if self.data is not "":
sql_database_table = SQL.fetch_all_class_data()
if sql_database_table == PermissionError:
self.my_view.user_has_no_file_permission()
elif sql_database_table == FileNotFoundError:
self.my_view.file_not_found_message()
elif sql_database_table == TypeError:
self.my_view.file_not_loaded_warning()
elif sql_database_table == AttributeError:
self.my_view.file_not_loaded_warning()
elif sql_database_table == Exception:
self.my_view.generic_error_message()
else:
self.my_view.read_database_file(sql_database_table)
else:
self.my_view.file_not_loaded_warning()
# Press 6 to load text file, convert data to PEP8 python format then convert file to pickle
# format in same directory
elif user_input == "6":
self.data = FileHandler.read_file()
if self.data == FileNotFoundError:
self.my_view.file_not_found_message()
else:
self.my_view.file_loaded_message()
self.prep_pep8()
pickle_status = Pickler.pickle_file(self.pep8_content)
if pickle_status == PermissionError:
self.my_view.user_has_no_file_permission()
elif pickle_status == FileNotFoundError:
self.my_view.file_not_found_message()
elif pickle_status == Exception:
self.my_view.generic_error_message()
else:
self.my_view.pickle_success_message()
# Press 7 to load data from pickle file
elif user_input == "7":
if self.data is not "":
pickle_content = Pickler.unpickle_file()
if pickle_content == PermissionError:
self.my_view.user_has_no_file_permission()
elif pickle_content == FileNotFoundError:
self.my_view.file_not_found_message()
elif pickle_content == Exception:
self.my_view.generic_error_message()
else:
self.my_view.print_my_pickle_content(pickle_content)
self.my_view.file_loaded_message()
else:
self.my_view.file_not_loaded_warning()
# Exit
elif user_input == "8":
incorrect_input = False
self.my_view.exit_program()
else:
self.my_view.user_has_wrong_input()
def get_option_type(option):
if option is None:
return None
return SQL().get_column_datatypes(column=option, singular=True)
def __upload_content(table_name, content):
return SQL().excel_to_sql(table_name=table_name,
excel_file=__parse_excel(content))
import pandas as pd
from config.important_attributes import attributes
from model.database import SQL
from model.dataframe import *
from controller.graph_components.regression import GraphMode
sql = SQL()
db_tables = sql.get_table_names()
row = {
u'namespace': u'dash_html_components',
u'type': u'Div',
u'props': {
u'className':
u'item-wrapper',
u'n_clicks_timestamp':
1532117798075L,
u'n_clicks':
1,
u'children': [{
u'namespace': u'dash_html_components',
u'type': u'Div',
u'props': {
u'className':
u'table-heading overflow-auto item-element-margin',
u'children': [{
u'namespace': u'dash_html_components',
u'type': u'H4',
u'props': {
u'className': u'header-title panel-left table-col-3',
def load_series_field(dump):
return [{
'label': table,
'value': table
} for table in SQL().get_table_names()]
def delete_button(click, container):
if click is not None and click != u'None':
table = container[0]['props']['children']
SQL().delete_table(table)
return
def load_table_container(dump):
return generate_table_container(SQL().get_table_names())
def test(dummy):
test_sql = SQL()
df = test_sql.get_table("testtable")
return generate_table(df)
def load_filter_specification(option):
# If nothing selected. Prevents loading empty filter options
if option is not None:
option_type = SQL().get_column_datatypes(column=option, singular=True)
return generate_filter_input(option_type)
def load_filter(dump):
return [{'label': i, 'value': i} for i in SQL().get_column_names()]
def obtain_data(dummy):
sql = SQL()
dff = sql.get_table(table_name)
return dff.to_json()
def update_graph(filtered_df_json, value, settings, graph_mode, clusters,
threshold, graphName, xLabel, yLabel, zLabel, extraMin,
extraMax, seriesInput, dbTableInput, figure, vessels,
*filter_settings):
if figure is not None:
figure['data'] = []
minSet = []
gformula = ""
gsols = 0.0
gr_squared = 0.0
valuesOrigin = ""
# Populate with 2D Data when X and Y set TODO: Remove hardcode + Account for 3D
if value[1] is None or value[2] is None:
figure['data'] = []
else:
# Create the dataset for the vessels selected
# Get specifications
specifications = []
for i in range(1, len(filter_settings), 3):
specifications.append(
get_condition(filter_settings[i], filter_settings[i + 1],
filter_settings[i + 2]))
# Cleanup and prepare conditions
conditions = []
for specification in specifications:
for condition in specification:
conditions.append(condition)
# Obtain filtered df
df = []
singleLineAll = False
secondAll = False
for vessel in filter_settings[0]:
if vessel == "All" and 'multiline' not in settings:
df = SQL().get_df_from_series(dbTableInput, seriesInput)
singleLineAll = True
break
elif vessel == "All":
secondAll = True
continue
df.append(dfs[vessel].get_filtered(conditions=conditions))
if len(df) == 0 and secondAll:
singleLineAll = True
df = SQL().get_df_from_series(dbTableInput, seriesInput)
if singleLineAll:
dfsDF = df
else:
dfsDF = pd.concat(df)
# Remove any NaN values
print("THIS IS VALUE: {}".format(value))
if value[0] == "2D":
dfsDF = dfsDF.dropna(subset=[value[1], value[2]])
else:
dfsDF = dfsDF.dropna(subset=[value[1], value[2], value[3]])
if threshold != "None":
# Remove outliers NOTE: Adjust the threshold to modify how strictly filtered the data will be. So far tested 1, 1.5, 3. Strict ~ Lax
mean = np.mean(dfsDF[value[1]])
stdio = np.std(dfsDF[value[1]])
print "Mean: " + str(mean) + " Std: " + str(stdio)
dfsDF = dfsDF[np.abs(dfsDF[value[1]] - mean) <= (threshold *
stdio)]
mean = np.mean(dfsDF[value[2]])
stdio = np.std(dfsDF[value[2]])
print "Mean: " + str(mean) + " Std: " + str(stdio)
dfsDF = dfsDF[np.abs(dfsDF[value[2]] - mean) <= (threshold *
stdio)]
if value[0] == "3D":
mean = np.mean(dfsDF[value[3]])
stdio = np.std(dfsDF[value[3]])
print "Mean: " + str(mean) + " Std: " + str(stdio)
dfsDF = dfsDF[np.abs(dfsDF[value[3]] -
mean) <= (threshold * stdio)]
# unqVessels = dfsDF['Vessel'].unique().tolist()
unqVessels = filter_settings[0]
print unqVessels
# Set axis labels if any
if xLabel == "":
xName = value[1]
else:
xName = xLabel
if yLabel == "":
yName = value[2]
else:
yName = yLabel
if graphName == "":
if value[0] == "2D":
gName = value[1] + " vs " + value[2]
else:
gName = value[1] + " vs " + value[2] + " vs " + value[3]
else:
gName = graphName
# Multiline Logic
if 'multiline' in settings:
counter = 0
benchmark = 0
annotation = []
sfList = SQL().get_vessel_codes()
for vessel in unqVessels:
global valuesOrigin
valuesOrigin = vessel
print list(dfsDF)
if vessel == "All":
vesselRow = SQL().get_df_from_series(
dbTableInput, seriesInput)
if value[0] == "2D":
vesselRow = vesselRow.dropna(
subset=[value[1], value[2]])
else:
vesselRow = vesselRow.dropna(
subset=[value[1], value[2], value[3]])
else:
vesselRow = dfsDF.loc[dfsDF['Vessel'] == vessel]
# Clustering & Hover Data Generation
hoverData = []
if 'clustering' in settings:
vesselRow = k_means(value, vesselRow, clusters)
vesselRow['Vessel'] = pd.Series(vessel,
index=vesselRow.index)
hoverData = np.c_[vesselRow[value[1]],
vesselRow[value[2]]]
else:
# Generate the Hover Data
# Iterate each row from db
for key, value1 in vesselRow.iterrows():
placeholderText = ""
# Iterate each column in the row
for index, row in value1.items():
# Compare the value in important_attributes
for col in full_attributes:
if col == index:
if isinstance(row, float):
placeholderText += "<b>" + index + "</b>: " + str(
round(row, 3)) + "<br>"
else:
placeholderText += "<b>" + index + "</b>: " + str(
row) + "<br>"
break
hoverData.append(placeholderText)
if 'datapoints' in settings:
if value[0] == "2D":
scatterPoints = go.Scatter(
x=vesselRow[value[1].encode('utf8')],
y=vesselRow[value[2].encode('utf8')],
name=vessel + " Marker",
mode='markers',
marker=go.Marker(color=colorList[counter]),
text=hoverData,
)
else:
scatterPoints = go.Scatter3d(
x=vesselRow[value[1].encode('utf8')],
y=vesselRow[value[2].encode('utf8')],
z=vesselRow[value[3].encode('utf8')],
name=vessel + " Marker",
mode='markers',
marker=go.Marker(color=colorList[counter]),
text=hoverData,
)
figure['data'].append(scatterPoints)
if 'regression' in settings:
if value[0] == "2D":
line_data, r_squared, sols, formula = regression(
vesselRow[value[1].encode('utf8')],
vesselRow[value[2].encode('utf8')], graph_mode,
extraMin, extraMax)
# tmpLst = []
# tmpLst.append(r_squared)
# tmpLst.append(sols)
# tmpLst.append(formula)
# gformula[vessel] = tmpLst
# gformula[vessel] = [r_squared, sols, formula]
global gr_squared, gsols, gformula
gr_squared = r_squared
gsols = sols
gformula = formula
eqString, supScript = generateEquationString(
formula)
bestFit = go.Scatter(
x=line_data['x'],
y=line_data['y'],
name=vessel + ' Line',
mode='lines',
marker=go.Marker(color=colorList[counter]),
)
figure['data'].append(bestFit)
if benchmark == 0:
benchmark = max(line_data['y'])
if vessel == "All":
vslCde = "All"
else:
vslCde = sfList[vessel]
annotation.append(
go.Annotation(x=min(line_data['x']) + 10,
y=benchmark -
counter * benchmark * 0.1,
text=vslCde + ": " +
eqString.format(*supScript),
showarrow=False))
layout2d = go.Layout(
title=gName,
plot_bgcolor='rgb(229, 229, 229)',
xaxis=go.XAxis(
title=xName,
zerolinecolor='rgb(255,255,255)',
gridcolor='rgb(255,255,255)'),
yaxis=dict(title=yName,
zerolinecolor='rgb(255,255,255)',
gridcolor='rgb(255,255,255)'),
annotations=annotation,
# yaxis2=dict(title='Percentage', gridcolor='blue', overlaying='y', side='right', range=[100,0]),
)
figure['layout'] = layout2d
else:
surfacePlot, surfaceLayout = plot_3d(
vesselRow[value[1].encode('utf8')],
vesselRow[value[2].encode('utf8')],
vesselRow[value[3].encode('utf8')], value[1],
value[2], value[3])
figure['data'].append(surfacePlot)
figure['layout'] = surfaceLayout
counter += 1
else: # If no multiline
# Clustering & Hover Data Generation
hoverData = []
if 'clustering' in settings:
dfsDF = k_means(value, dfsDF, clusters)
hoverData = np.c_[dfsDF[value[1]], dfsDF[value[2]]]
else:
# Generate the Hover Data
# Iterate each row from db
for key, value1 in dfsDF.iterrows():
placeholderText = ""
# Iterate each column in the row
for index, row in value1.items():
# Compare the value in important_attributes
for col in full_attributes:
if col == index:
if isinstance(row, float):
placeholderText += "<b>" + index + "</b>: " + str(
round(row, 3)) + "<br>"
else:
placeholderText += "<b>" + index + "</b>: " + str(
row) + "<br>"
break
hoverData.append(placeholderText)
if value[0] == "2D":
# Add scatter from data set if 'datapoints' toggled
if len(figure['data']) < 1:
figure['data'].append({})
if 'datapoints' in settings:
figure['data'][0] = go.Scatter(
x=dfsDF[value[1].encode('utf8')],
y=dfsDF[value[2].encode('utf8')],
name='Data Marker',
mode='markers',
text=hoverData,
)
else:
figure['data'][0] = None
# Add Line/Curve if 'regression' toggled
if len(figure['data']) < 2:
figure['data'].append({})
if 'regression' in settings:
line_data, r_squared, sols, formula = regression(
dfsDF[value[1].encode('utf8')],
dfsDF[value[2].encode('utf8')], graph_mode,
extraMin, extraMax)
print "R-Squared: " + str(r_squared)
print "Sum of Least Squares: " + str(sols)
print "A Formula: "
print formula
# global gr_squared, gsols, gformula
gr_squared = r_squared
gsols = sols
gformula = formula
eqString, supScript = generateEquationString(formula)
figure['data'][1] = go.Scatter(
x=line_data['x'],
y=line_data['y'],
name='Line',
mode='lines',
)
annotation = go.Annotation(x=min(line_data['x']) + 10,
y=max(line_data['y']),
text="y=" +
eqString.format(*supScript),
showarrow=False)
layout2d = go.Layout(
title=gName,
plot_bgcolor='rgb(229, 229, 229)',
xaxis=go.XAxis(title=xName,
zerolinecolor='rgb(255,255,255)',
gridcolor='rgb(255,255,255)'),
yaxis=dict(title=yName,
zerolinecolor='rgb(255,255,255)',
gridcolor='rgb(255,255,255)'),
annotations=[annotation],
# yaxis2=dict(title='Percentage', gridcolor='blue', overlaying='y', side='right', range=[100,0]),
)
figure['layout'] = layout2d
else:
figure['data'][1] = None
else:
# 3D
# Add scatter from data set if 'datapoints' toggled
if len(figure['data']) < 1:
figure['data'].append({})
if 'datapoints' in settings:
figure['data'][0] = go.Scatter3d(
x=dfsDF[value[1].encode('utf8')],
y=dfsDF[value[2].encode('utf8')],
z=dfsDF[value[3].encode('utf8')],
name='Data Marker',
mode='markers',
text=hoverData,
)
else:
figure['data'][0] = None
# Add Line/Curve if 'regression' toggled
if len(figure['data']) < 2:
figure['data'].append({})
if 'regression' in settings:
surfacePlot, surfaceLayout, minimumSet = plot_3d(
dfsDF[value[1].encode('utf8')],
dfsDF[value[2].encode('utf8')],
dfsDF[value[3].encode('utf8')], value[1], value[2],
value[3])
figure['data'][1] = surfacePlot
figure['layout'] = surfaceLayout
minSet = minimumSet
print "MINSET"
print minSet
else:
figure['data'][1] = None
figure['layout']['scene']['xaxis']['title'] = xName
figure['layout']['scene']['yaxis']['title'] = yName
figure['layout']['scene']['zaxis']['title'] = zName
figure['layout']['title'] = gName
# Clean figure data
figure['data'] = [i for i in figure['data'] if i is not None]
return figure
return default_figure
0, # 2. Sea State
0, # 3. Swell
10, # 4. Speed
14, # 5. RPM
34, # 6. Trim
0, # 7. Power
]
# Display column names with index no.
def get_names(names):
for name in names:
print "{}. {}".format(names.index(name), name)
sql = SQL()
# # Read Attributes
# get_names(attributes)
# Get database tables
db_tables = sql.get_table_names()
# get_names(db_tables)
# Get column names for given database
column_names = sql.get_column_names(db_tables[0])
# get_names(column_names)
# Insert Speed
for i in range(len(col_no)):
if col_no[i] > 0:
def update_filer(value, mode):
if value > 0:
options = [{'label': i, 'value': i} for i in SQL().get_column_names()]
return generate_graph(mode, options)
def get_table(dummy):
sql = SQL()
df = sql.get_table('testtable')
return generate_table(df)