def display(self):
plt.figure()
for car in self.cars:
node0, node1 = car.route[0]
if car.arrive_to_destination():
plt.plot(
((nodes[node0]['longitude'] + nodes[node1]['longitude']) /
2),
((nodes[node0]['latitude'] + nodes[node1]['latitude']) /
2), 'gs')
else:
plt.plot(
((nodes[node0]['longitude'] + nodes[node1]['longitude']) /
2),
((nodes[node0]['latitude'] + nodes[node1]['latitude']) /
2), 'bs')
for system in self.systems:
node0, node1 = system.route[0]
plt.plot(
((nodes[node0]['longitude'] + nodes[node1]['longitude']) / 2),
((nodes[node0]['latitude'] + nodes[node1]['latitude']) / 2),
'rs')
mplleaflet.save_html(
fileobj='images/{:02d}{:02d}{:02d}_simulation_{}.html'.format(
self.time_hours_from_beginning,
self.time_minutes_from_beginning,
self.time_seconds_from_beginning, self.simulation_id))
def leafletdo(db):
uids = []
lats = []
longs = []
tokens = []
users = db.child("users").get().val()
for key, value in users.items():
# print(key, value)
if "location" in value and "token" in value:
uids.append(key)
lats.append(float(value["location"]["lat"]))
longs.append(float(value["location"]["long"]))
tokens.append(value["token"]["token"])
print(uids)
# plt.figure(figsize=(8,6))
# fig = plt.figure()
fig, ax = plt.subplots(figsize=(8, 8))
for i, uid in enumerate(uids):
x = longs[i]
y = lats[i]
ax.scatter(x, y, marker='x', color='red')
ax.text(x + 0.3, y + 0.3, uid, fontsize=9)
#plt.figure(figsize=(8,6))
#fig = plt.figure()
#plt.plot(longs, lats, 'ro')
mplleaflet.save_html(fileobj=os.path.join(dir_path, "templates/map.html"),
fig=fig)
def bikesplorela():
"""Main page"""
form = LoginForm()
if form.validate_on_submit():
session['pointA'] = form.pointA.data
session['pointB'] = form.pointB.data
geo_df = gs.read_file('app/static/collisions.geojson')
locs = {
'Loc1': [-118.336389, 34.100833],
'Loc2': [-118.356, 34.0628],
'Loc3': [-118.3208331, 34.0837995]
}
Loc1 = form.pointA.data
Loc2 = form.pointB.data
x1, y1 = locs[form.pointA.data]
x2, y2 = locs[form.pointB.data]
reqtxt = 'http://router.project-osrm.org/route/v1/driving/{x1},{y1};{x2},{y2}?overview=simplified&exclude=motorway&annotations=true&geometries=geojson'.format(
x1=x1, y1=y1, x2=x2, y2=y2)
rout = requests.get(reqtxt)
route = json.loads(rout.text)
coords = route['routes'][0]['geometry']['coordinates']
xs_route = [c[0] for c in coords]
ys_route = [c[1] for c in coords]
map_file = 'app/static/map_{loc1}_{loc2}.html'.format(loc1=Loc1,
loc2=Loc2)
map_file2 = 'static/map_{loc1}_{loc2}.html'.format(loc1=Loc1,
loc2=Loc2)
f, ax = plt.subplots(1)
geo_df.geometry.plot(alpha=0.9, edgecolor='red',
ax=ax) #plot dangergrid
ax.plot(xs_route, ys_route, color='blue', linewidth=4, alpha=0.5)
mplleaflet.save_html(fig=f, crs=geo_df.crs, fileobj=map_file)
#fname = str(tmp.communicate()[0].strip(b'\n'),'utf-8')
elif request.method == 'GET':
form.pointA.data = session.get('pointA', '')
form.pointB.data = session.get('pointB', '')
map_file2 = 'static/map.html'
return render_template('bikesplorela.html', form=form, map_file=map_file2)
def _plot_mplleafflet(lats, lons, observable_values_in_minutes,
observable_name, scalar_mappable, basename, node_names):
import mplleaflet
fig = plt.figure()
ax = fig.add_subplot(111)
colors = scalar_mappable.to_rgba(observable_values_in_minutes)
assert (isinstance(ax, matplotlib.axes.Axes))
ax.scatter(lons, lats, c=colors, edgecolors=colors, s=10)
cbar = fig.colorbar(scalar_mappable)
ax.set_title(observable_name)
mplleaflet.save_html(fig, basename + observable_name + ".html")
def leaflet_heatmap(cls, yy, xx, Z, bounds, tiles='cartodb_positron', map_path='_map.html', legend_text='',
description_title='', description_text=''):
"""
Plots a heatmap over a leaflet layer
Parameters
----------
yy: np.array of y-coordinates
xx: np.array of x-coordinates
Z: np.array of Z-coordinates
bounds: ((x_min, y_min), (x_max, y_max)) or None
tiles: str
Name of the tiles to use
map_path: str
Where to output html
legend_text: str
Text to put on the legend
description_title: str
Title of the description box
description_text: str
Text used to describe the map
Returns
-------
"""
plt.figure()
plt.pcolormesh(yy, xx, Z, cmap='YlGnBu', alpha=1)
plt.xlim(bounds[0][0], bounds[1][0])
plt.ylim(bounds[0][1], bounds[1][1])
vmin = Z.min()
vmax = Z.max()
if tiles == 'cartodb_positron':
tiles = 'http://{s}.basemaps.cartocdn.com/light_all/{z}/{x}/{y}.png'
attr = '© <a href="http://cartodb.com/attributions">CartoDB</a> | ' \
'© <a href="http://www.openstreetmap.org/copyright">OpenStreetMap</a> | ' \
'© Michael Clawar & Raaid Arshad'
tiles = (tiles, attr)
elif tiles == 'osm_mapnik':
tiles = 'http://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png'
attr = '© <a href="http://www.openstreetmap.org/copyright">OpenStreetMap</a> | ' \
'© Michael Clawar & Raaid Arshad</a>'
tiles = (tiles, attr)
else:
raise ValueError
mplleaflet.save_html(fileobj=map_path, tiles=tiles)
cls.add_legend_mplleaflet(map_path, legend_text=legend_text, vmin=vmin, vmax=vmax,
description_title=description_title, description_text=description_text)
def makeMap(self, linewidth=2.5, output_file='map'):
for axarr in self.axarr:
axarr.lines = []
if self.map:
raise TrackException('Map background found in the figure', 'Remove it to create an interactive HTML map.')
if 'Color' in self.track_df.df:
for point, next_point in self.computePoints(linewidth=linewidth):
pass
else:
self.computeTracks(linewidth=linewidth)
mplleaflet.save_html(fig=self.fig, tiles='esri_aerial', fileobj=output_file + '.html')#, close_mpl=False) # Creating html map
def save(fig=None, path='_map.html', **kwargs):
"""
Convert a Matplotlib Figure to a Leaflet map. Open in a browser
Parameters
----------
fig : figure, default gcf()
Figure used to convert to map
path : string, default '_map.html'
Filename where output html will be saved
"""
fullpath = os.path.abspath(path)
with open(fullpath, 'w') as f:
mplleaflet.save_html(fig, fileobj=f, **kwargs)
def plot_train(trip, data_date, route):
print trip
data_path = 'data/user_data/' + data_date
u = 'u_' + trip[0]
f = open(os.path.join(data_path, u), "r")
data = [[row[0], int(row[4]),
float(row[2]), float(row[3])]
for row in csv.reader(f, delimiter='|')
] #[imsi, unix_time, lon, lat]
data_df = pandas.DataFrame(data, columns=['imsi', 'time', 'lon', 'lat'])
data_df = data_df.sort_values(['time'])
data = data_df.values.tolist()
#data = np.array([row.rstrip().split("|") for row in f]) #-1943160471|16:57:03,966,490|121.526|25.088|1483520223
f.close()
start = int(trip[1])
end = int(trip[2])
user = []
for row in data:
if (row[1] >= start) and (row[1] <= end):
user.append([row[2], row[3]])
print user
### plot user data ###
user = np.array(user)
plt.hold(True)
plt.plot(user[:, 0], user[:, 1], 'mo', ms=15)
plt.plot(user[:, 0], user[:, 1], 'r--', linewidth=5)
### plot train station ###
rail_station = np.array(
[[rail[station]['position'][0], rail[station]['position'][1]]
for station in trip[3:5]])
print rail_station
plt.plot(rail_station[:, 0], rail_station[:, 1], 'bo', ms=15)
try:
os.makedirs('plot_train/' + data_date)
except OSError as e:
if e.errno != errno.EEXIST:
raise
mplleaflet.save_html(fileobj="./plot_train/%s/%s_%s_%s.html" %
(data_date, u, trip[1], trip[2]))
print
def myshow(fig=None, path='_map.html', **kwargs):
"""
Convert a Matplotlib Figure to a Leaflet map. Open in a browser
Parameters
----------
fig : figure, default gcf()
Figure used to convert to map
path : string, default '_map.html'
Filename where output html will be saved
See fig_to_html() for description of keyword args.
"""
import webbrowser
import os
fullpath = os.path.abspath(path)
with open(fullpath, 'w') as f:
mplleaflet.save_html(fig, fileobj=f, **kwargs)
def draw_map():
stations = get_stations()
stations = pd.json_normalize(stations)
longitude = stations["lon"]
latitude = stations["lat"]
status = get_status()
status = pd.json_normalize(status)
available_bikes = status["num_bikes_available"]
fig = plt.figure()
plt.scatter(longitude,
latitude,
s=stations["capacity"] * 3,
c=available_bikes,
cmap="RdYlGn",
vmin=0,
vmax=10)
for lon, lat, no_available in zip(longitude, latitude, available_bikes):
label = str(no_available)
plt.annotate(label, (lon, lat))
mplleaflet.save_html(fileobj="static/map/map.html")
def plot_map(self):
'''
Utility function that plots a route on a map
Returns
-------
None.
'''
fig, ax = plt.subplots()
df = self.map_df
df = df.dropna()
ax.plot(df['lon'],
df['lat'],
color='darkorange',
linewidth=5,
alpha=0.5)
mplleaflet.save_html(fig, fileobj="map.html")
opener = "xdg-open"
x = os.fork()
if (x == 0):
subprocess.call([opener, "map.html"])
def gen_map_html(OUTPUT_HTML_FILE):
"""
visualize using mplleaflet
Map data can be downloaded at: mapzen/metro-extracts, https://www.interline.io/osm/extracts/
in imposm-geojson format.
:return:
"""
# overlay map
admin = gpd.GeoDataFrame.from_file('singapore.imposm-geojson/singapore_admin.geojson')
singapore = admin.iloc[0]
print(singapore)
print("done")
# draw base map
gpd.GeoSeries(singapore.geometry).plot()
# places of interest
places = gpd.GeoDataFrame.from_file('singapore.imposm-geojson/singapore_corona.geojson')
print(places)
print("done")
# bounded inside the map
places = places[places.geometry.within(singapore.geometry)]
print(places)
# draw places
ax = places.geometry.plot(color='red')
#### Output
#mplleaflet.display(fig=ax.figure) # To display it right at the notebook.
#mplleaflet.show(fig=ax.figure) # To output _map.html file and display it in your browser.
mplleaflet.save_html(fig=ax.figure, fileobj=OUTPUT_HTML_FILE)
def leaflet_heatmap(cls,
yy,
xx,
Z,
bounds,
tiles='cartodb_positron',
map_path='_map.html',
legend_text='',
description_title='',
description_text=''):
"""
Plots a heatmap over a leaflet layer
Parameters
----------
yy: np.array of y-coordinates
xx: np.array of x-coordinates
Z: np.array of Z-coordinates
bounds: ((x_min, y_min), (x_max, y_max)) or None
tiles: str
Name of the tiles to use
map_path: str
Where to output html
legend_text: str
Text to put on the legend
description_title: str
Title of the description box
description_text: str
Text used to describe the map
Returns
-------
"""
plt.figure()
plt.pcolormesh(yy, xx, Z, cmap='YlGnBu', alpha=1)
plt.xlim(bounds[0][0], bounds[1][0])
plt.ylim(bounds[0][1], bounds[1][1])
vmin = Z.min()
vmax = Z.max()
if tiles == 'cartodb_positron':
tiles = 'http://{s}.basemaps.cartocdn.com/light_all/{z}/{x}/{y}.png'
attr = '© <a href="http://cartodb.com/attributions">CartoDB</a> | ' \
'© <a href="http://www.openstreetmap.org/copyright">OpenStreetMap</a> | ' \
'© Michael Clawar & Raaid Arshad'
tiles = (tiles, attr)
elif tiles == 'osm_mapnik':
tiles = 'http://{s}.tile.openstreetmap.org/{z}/{x}/{y}.png'
attr = '© <a href="http://www.openstreetmap.org/copyright">OpenStreetMap</a> | ' \
'© Michael Clawar & Raaid Arshad</a>'
tiles = (tiles, attr)
else:
raise ValueError
mplleaflet.save_html(fileobj=map_path, tiles=tiles)
cls.add_legend_mplleaflet(map_path,
legend_text=legend_text,
vmin=vmin,
vmax=vmax,
description_title=description_title,
description_text=description_text)
def save(fig=None, path='_map.html', template='base.html' ,**kwargs):
fullpath = os.path.abspath(path)
with open(fullpath, 'w') as f:
mplleaflet.save_html(fig, fileobj=f, template =template, **kwargs)
troop_positions = pd.read_fwf("../../data/napoloen/troops.txt")
troop_positions = gpd.GeoDataFrame(data=troop_positions,
geometry=troop_positions\
.apply(lambda srs: Point(srs['long'], srs['lat']),
axis='columns'))
subsrs = []
for a, b in zip(range(len(troop_positions) - 1), range(1, len(troop_positions))):
srs = troop_positions.iloc[b]
srs = srs.rename({'geometry': 'from'})
srs['to'] = troop_positions.iloc[a].geometry
subsrs.append(srs)
troop_movements = pd.concat(subsrs, axis=1).T
troop_movements = troop_movements[['survivors', 'direction', 'group', 'from', 'to']]
troop_movements['direction'] = troop_movements.direction.map(lambda d: 0 if d == 'A' else 1)
# Plot the data.
# We'll use a custom colormap, to match the one that Minard uses.
from matplotlib.colors import LinearSegmentedColormap
colors = [(215/255, 193/255, 126/255), (37/255, 37/255, 37/255)]
cm = LinearSegmentedColormap.from_list('minard', colors)
gplt.sankey(troop_movements, start='from', end='to',
scale='survivors', limits=(0.5, 45),
hue='direction', categorical=True, cmap=cm)
fig = plt.gcf()
mplleaflet.save_html(fig, fileobj='minard-napoleon-russia.html')
# Load the data (uses the `quilt` package).
import geopandas as gpd
from quilt.data.ResidentMario import geoplot_data
troop_movements = gpd.read_file(geoplot_data.napoleon_troop_movements())
troop_movements['from'] = troop_movements.geometry.map(lambda v: v[0])
troop_movements['to'] = troop_movements.geometry.map(lambda v: v[1])
# Plot the data. We'll use a custom colormap, to match the one that Minard uses.
import geoplot as gplt
import matplotlib.pyplot as plt
import mplleaflet
from matplotlib.colors import LinearSegmentedColormap
colors = [(215 / 255, 193 / 255, 126 / 255), (37 / 255, 37 / 255, 37 / 255)]
cm = LinearSegmentedColormap.from_list('minard', colors)
gplt.sankey(troop_movements,
start='from',
end='to',
scale='survivors',
limits=(0.5, 45),
hue='direction',
categorical=True,
cmap=cm)
fig = plt.gcf()
mplleaflet.save_html(fig, fileobj='minard-napoleon-russia.html')
def main():
##### READ IN COMMAND LINE ARGUMENTS
parser = argparse.ArgumentParser(
description="""
Purpose: Script to calculate validation metrics for EASOS model output with respect to observations from
satellite measurements or coastal reports. Minimum input requirement is a pair of GeoJSON files valid at a given time instance,
one containing the validation data (either from satellites or coastal reports) and the other the corresponding model prediction data.
Each GeoJSON file should contain a geodataframe with geometries that define the oil spill area/extent.
Both deterministic (i.e. 'Best Estimate') and probabilistic model output are supported. """,
epilog=
"Example of use: ./Calc_2D_MOE_GeoJSON.py <obsFile> <modelFile> <modelType> <valType> [--noOilFile NOOILFILE] [--crs CRS] ",
)
parser.add_argument(
"obsFile",
help=
"Required. Absolute or relative path to observation data file in GeoJSON format",
type=str,
)
parser.add_argument(
"modelFile",
help=
"Required. Absolute/relative path to model output file (either deterministic or probabilistic) in GeoJSON format",
type=str,
)
parser.add_argument(
"modelType",
help=
"Required. Type of model output, either 'BE' (best estimate, aka deterministic) or 'Prob' (probabilistic). Model contours \
are assumed to be cut-outs, i.e. no overlap with contours of a higher level.",
type=str,
)
parser.add_argument(
"valType",
help=
"Required. Type of validation to be performed, either 'Satellite' or 'Coastal'",
type=str,
)
parser.add_argument(
"--noOilFile",
help=
"Optional path to a file in GeoJSON format defining the region where oil was not observed in the satellite image. \
If this is specified, any model output that lies outside this detection region will be excluded from the analysis.",
type=str,
)
parser.add_argument(
"--crs",
help=
"Optional integer specifying the crs code to convert obs and model data to. Default value is 3857. \
See http://epsg.io/3857 for more details",
type=int,
default=3857,
)
args = parser.parse_args()
obsFile = args.obsFile
modelFile = args.modelFile
modelType = args.modelType
valType = args.valType
noOilFile = args.noOilFile
crs = args.crs
#####
##### READ IN GEOJSON FILES, CHECK VALIDITY, AND RETURN AS GEODATAFRAMES
oil, model, no_oil, casename, time, plevs = read_geojson(
obsFile, modelFile, noOilFile, modelType, valType, crs)
if valType == "Coastal":
# Do a basic plot of the model coastal prediction with the obs regions highlighted and save as a png file
modelplot, ax = plot_coastal_maps(oil,
model,
casename,
time,
modelType,
noOil=no_oil,
levels=plevs)
modelplot.savefig(
"/media/Coastal_map_" + str(casename) + "_" + str(modelType) +
"_" + str(time) + ".png",
bbox_inches="tight",
)
if modelType == "BE":
leaf.save_html(
fig=ax.figure,
fileobj="/media/Interactive_map_" + str(casename) + "_" +
str(modelType) + "_" + str(time) + ".html",
)
#####
##### PREPARE AND UPDATE GEODATAFRAMES WITH OBS AREA, MODEL AREA AND OVERLAP AREA
oil, model_known, overlap, plevs = calc_poly_overlap(
oil, model, no_oil, casename, time, noOilFile, modelType, valType, crs)
if overlap.empty:
Aob = oil["obs_area"]
Apr = model_known["area_full_contour"]
print("Overlap geodataframe is empty; skipping 2-D MOE calculation")
else:
if valType == "Satellite":
# Produce a basic map of the obs, model and overlap regions and save in png format
modelplot = plot_area_maps(oil,
model_known,
overlap,
casename,
time,
modelType,
levels=plevs)
modelplot.savefig(
"/media/Area_maps_" + str(casename) + "_" + str(modelType) +
"_" + str(time) + ".png",
bbox_inches="tight",
)
##### CALCULATE THE 2-D MEASURE OF EFFECTIVENESS AND GENERATE PLOTS
##### For details, see Warner et al 2004., J. Appl. Met
Aob = overlap["obs_area"]
Apr = overlap["area_full_contour"]
Aov = overlap["overlap_full_contour"]
# Call function to return x and y components of the 2-D MOE
(x, y) = calc_2DMOE(Aob, Apr, Aov)
# Generate 2-D MOE space diagram and save in png format
olevs = (overlap.contourlev).to_numpy()
if modelType == "BE":
MOEfig = plot_2D_MOE_scat(x, y, modelType, casename, time)
elif modelType == "Prob":
MOEfig = plot_2D_MOE_scat(x,
y,
modelType,
casename,
time,
levels=olevs)
MOEfig.savefig(
"/media/2D_MOE_" + str(casename) + "_" + str(modelType) + "_" +
str(valType) + "_" + str(time) + ".png",
bbox_inches="tight",
)
#####
##### FOR SATELLITE VALIDATION AGAINST DETERMINISTIC OUTPUT, CALCULATE ADDITIONAL SKILL SCORES
if valType == "Satellite" and modelType == "BE":
from calc_metrics import calc_area_ss, calc_centroid_ss
from plot_maps_metrics import plot_ss_scat, plot_centroid_map
print("Area (in km^2) of observed spill is : ", Aob[0])
print("Area (in km^2) of modelled spill is : ", Apr[0])
# Call function to calcluate Area Skill Score, Ass
Ass = calc_area_ss(Aob[0], Apr[0])
print("Area skill score is : ", Ass)
# Call function to calculate Centroid Skill Score, Css
Css, obs_centroid, model_centroid, minpoint, maxpoint = calc_centroid_ss(
oil, model_known)
print("Centroid skill score is : ", Css)
# Plot the modelled and observed oil spill areas, with centroids and distances indicated
centroidfig = plot_centroid_map(
oil,
obs_centroid,
model_known,
model_centroid,
minpoint,
maxpoint,
casename,
time,
)
centroidfig.savefig(
"/media/Centroid_map_" + str(casename) + "_" + str(time) + ".png",
bbox_inches="tight",
)
# Proceed to plot skill score results on a scatter diagram
SSfig = plot_ss_scat(Ass, Css, casename, time)
SSfig.savefig(
"/media/Skillscores_scatterplot_" + str(casename) + "_" +
str(time) + ".png",
bbox_inches="tight",
)
overlay = nx.read_gml(overlay_path).to_undirected()
mapping = {}
for ii, node in enumerate(underlay.nodes()):
mapping[str(ii)] = node
overlay = nx.relabel_nodes(overlay, mapping).to_undirected()
fig, ax = plt.subplots()
nx.draw_networkx_nodes(overlay,
pos=pos_dict,
node_size=10,
node_color='red',
edge_color='k',
alpha=.5,
with_labels=True)
nx.draw_networkx_edges(overlay,
pos=pos_dict,
edge_color='blue',
alpha=1,
width=5.0)
nx.draw_networkx_labels(overlay, pos=pos_dict, label_pos=10.3)
mplleaflet.display(fig=ax.figure)
mplleaflet.save_html(fig=ax.figure,
fileobj=os.path.join(
"results", args.underlay,
"{}.html".format(args.architecture)))
if grppviz.has_edge(n1, n2):
grppviz[n1][n2][0]['linewidth'] += 2
grppviz[n1][n2][0]['cnt'] += 1
else:
grppviz.add_edge(n1, n2, linewidth=2.5)
grppviz[n1][n2][0]['color_st'] = 'black' if g_t.has_edge(
n1, n2) else 'red'
grppviz[n1][n2][0]['cnt'] = 1
grppviz.add_node(n1, lat=g.node[n1]['lat'], lon=g.node[n1]['lon'])
grppviz.add_node(n2, lat=g.node[n2]['lat'], lon=g.node[n2]['lon'])
for e in grppviz.edges(data=True):
e[2]['color_cnt'] = color_seq[1] if 'cnt' not in e[2] else color_seq[
e[2]['cnt']]
fig, ax = plt.subplots(figsize=(1, 10))
pos = {
k: (grppviz.node[k]['lon'], grppviz.node[k]['lat'])
for k in grppviz.nodes()
}
e_width = [e[2]['linewidth'] for e in grppviz.edges(data=True)]
e_color = [e[2]['color_cnt'] for e in grppviz.edges(data=True)]
nx.draw_networkx_edges(grppviz,
pos,
width=e_width,
edge_color=e_color,
alpha=0.7)
mplleaflet.save_html(fig, 'maps/solutions.html', tiles='cartodb_positron')
def plot_bus(trip, data_date, route):
# trip: [-1003216209,1483537823,1483539806,NWT16595,1,11,42]
# trip: [imsi,start_time,end_time,bus_number,go_or_back,start_stop,end_stop]
print trip
data_path = 'data/user_data/' + data_date
u = 'u_' + trip[0]
f = open(os.path.join(data_path, u), "r")
data = [[row[0], int(row[4]),
float(row[2]), float(row[3])]
for row in csv.reader(f, delimiter='|')
] #[imsi, unix_time, lon, lat]
data_df = pandas.DataFrame(data, columns=['imsi', 'time', 'lon', 'lat'])
data_df = data_df.sort_values(['time'])
data = data_df.values.tolist()
#data = np.array([row.rstrip().split("|") for row in f]) #-1943160471|16:57:03,966,490|121.526|25.088|1483520223
f.close()
'''
year = int(data_date[0:4])
month = int(data_date[5])
date = int(data_date[7])
### sort user data by datetime ###
for n,row in enumerate(data,0):
hour,minute,second = row[1].split(",")[0].split(":")
unix_time = int(time.mktime(time.strptime('%d-%d-%d %s:%s:%s'%(year,month,date,hour,minute,second), '%Y-%m-%d %H:%M:%S')))
data[n][4] = str(unix_time)
data = sorted(data,key=lambda x:x[4])
data = [[row[0],int(row[4]),float(row[2]),float(row[3])] for row in data] #imsi,unixtime,lon,lat
'''
start = int(trip[1])
end = int(trip[2])
user = []
for row in data:
if (row[1] >= start) & (row[1] <= end):
user.append([row[2], row[3]])
print user
### plot user data ###
user = np.array(user)
plt.hold(True)
plt.plot(user[:, 0], user[:, 1], 'go', ms=15)
plt.plot(user[:, 0], user[:, 1], 'r--', linewidth=5)
### plot route data ###
route_name = trip[3] + "_" + trip[4]
route_list = route[route_name][int(trip[5]):int(trip[6]) +
1] #from start_stop to end_stop
bus_line = np.array([[float(item[2]), float(item[3])]
for item in route_list]) #lon,lat => bus route
plt.plot(bus_line[:, 0], bus_line[:, 1], 'co', ms=15)
plt.plot(bus_line[:, 0], bus_line[:, 1], 'b--', linewidth=5)
try:
os.makedirs('plot_bus/' + data_date)
except OSError as e:
if e.errno != errno.EEXIST:
raise
mplleaflet.save_html(fileobj="./plot_bus/%s/%s_%s_%s.html" %
(data_date, u, trip[1], trip[2]))
print
