# In this script, we will analyze Heatmaps using plotly functions
import plotly.offline as pyo
import plotly.graph_objs as go
import pandas as pd
#dataset read
df = pd.read_csv('2010SantaBarbaraCA.csv')
# Heatmap data
data = [go.Heatmap(x=df['DAY'], #x - xaxis, y- yaxis and z- color axis
y=df['LST_TIME'],
z=df['T_HR_AVG'].values.tolist())] # for color axis, we need a lsit so we convert with .values.tolist() from column to list
layout = go.Layout(title='Heatmap for Temperature in Santa Barbara 2010')
fig=go.Figure(data=data,layout=layout)
pyo.plot(fig,filename='Basic-Heatmap-Temps.html')
#-------------------------------------------------------------------------------------------------------------------------------------------------#
#Different color scales
data1 = [go.Heatmap(x=df['DAY'], #x - xaxis, y- yaxis and z- color axis
y=df['LST_TIME'],
z=df['T_HR_AVG'].values.tolist(),
colorscale='Jet')]
layout1 = go.Layout(title='Heatmap for Temperature in Santa Barbara 2010 - Jet colormap')
def plot_heatmap(
df_corr,
colorscale="brwnyl",
cut_in_half=False,
showscale=True,
textfont="default",
xgap=1,
ygap=1,
transparent=True,
):
if textfont == "default":
textfont = {
"color": "#000000",
"size": 14,
"family": "Times New Roman"
}
trace = go.Heatmap(
colorscale=colorscale,
hoverinfo="text",
xgap=xgap,
ygap=ygap,
showscale=showscale,
)
if cut_in_half:
corr_triangle = np.array([[None for k in range(df_corr.shape[1] - 1)]
for j in range(df_corr.shape[0] - 1)])
for k, vals in enumerate(df_corr[1:].values):
corr_triangle[k][:k + 1] = np.round(vals[:k + 1], 2)
trace.update(
z=corr_triangle[::-1],
x=df_corr.index[:-1],
y=df_corr.index[1:][::-1],
text=corr_triangle[::-1],
)
fig = go.Figure(data=[trace])
annotations = []
for k, y in enumerate(df_corr.index):
for x in df_corr.index[:k]:
anot = go.layout.Annotation(
x=x,
y=y,
xref="x",
yref="y",
text=f"<b>{df_corr[x][y]: .2f}</b>",
showarrow=False,
font=textfont,
)
annotations.append(anot)
else:
trace.update(
z=df_corr.values[::-1],
x=df_corr.index,
y=df_corr.index[::-1],
text=np.round(df_corr.values[::-1], 2),
)
fig = go.Figure(data=[trace])
annotations = []
for k, y in enumerate(df_corr.index):
for x in df_corr.index:
anot = go.layout.Annotation(
x=x,
y=y,
xref="x",
yref="y",
text=f"<b>{df_corr[x][y]: .2f}</b>",
showarrow=False,
font=textfont,
)
annotations.append(anot)
fig.update_layout(showlegend=False, annotations=annotations)
if transparent:
fig.update_layout(plot_bgcolor="rgba(0,0,0,0)",
paper_bgcolor="rgba(0,0,0,0)")
return fig
import plotly.offline as pyo
import plotly.graph_objs as go
import pandas as pd
# Load CSV file from Datasets folder
df = pd.read_csv('../Datasets/Weather2014-15.csv')
# Preparing data
data = [
go.Heatmap(x=df['day'],
y=df['month'],
z=df['actual_max_temp'].values.tolist(),
colorscale='Jet')
]
# Preparing layout
layout = go.Layout(title='Max Temperature on Day of Week and Month of Year',
xaxis_title="Day of Week",
yaxis_title="Month")
# Plot the figure and saving in a html file
fig = go.Figure(data=data, layout=layout)
pyo.plot(fig, filename='heatmap.html')
Barcelona, 29/11/2012
"""
import argparse, os, sys
from datetime import datetime
import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
import plotly.plotly as py
import plotly.graph_objs as go
data = [
go.Heatmap(z=[[1, 20, 30, 50, 1], [20, 1, 60, 80, 30],
[30, 60, 1, -10, 20]],
x=['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday'],
y=['Morning', 'Afternoon', 'Evening'])
]
plot_url = py.plot(data, filename='labelled-heatmap')
def histplot(ax, x, handle, bins, title, xlab, ylab, xlog, ylog, \
normed=False, alpha=0.75):
"""
"""
#fig = plt.figure()
#ax = fig.add_subplot(111)
if xlog:
ax.set_xscale('log')
if ylog:
ax.set_yscale('log', nonposy='clip')
# the histogram of the data
def imshow(
img,
zmin=None,
zmax=None,
origin=None,
labels={},
x=None,
y=None,
animation_frame=None,
facet_col=None,
facet_col_wrap=None,
facet_col_spacing=None,
facet_row_spacing=None,
color_continuous_scale=None,
color_continuous_midpoint=None,
range_color=None,
title=None,
template=None,
width=None,
height=None,
aspect=None,
contrast_rescaling=None,
binary_string=None,
binary_backend="auto",
binary_compression_level=4,
binary_format="png",
text_auto=False,
):
"""
Display an image, i.e. data on a 2D regular raster.
Parameters
----------
img: array-like image, or xarray
The image data. Supported array shapes are
- (M, N): an image with scalar data. The data is visualized
using a colormap.
- (M, N, 3): an image with RGB values.
- (M, N, 4): an image with RGBA values, i.e. including transparency.
zmin, zmax : scalar or iterable, optional
zmin and zmax define the scalar range that the colormap covers. By default,
zmin and zmax correspond to the min and max values of the datatype for integer
datatypes (ie [0-255] for uint8 images, [0, 65535] for uint16 images, etc.). For
a multichannel image of floats, the max of the image is computed and zmax is the
smallest power of 256 (1, 255, 65535) greater than this max value,
with a 5% tolerance. For a single-channel image, the max of the image is used.
Overridden by range_color.
origin : str, 'upper' or 'lower' (default 'upper')
position of the [0, 0] pixel of the image array, in the upper left or lower left
corner. The convention 'upper' is typically used for matrices and images.
labels : dict with str keys and str values (default `{}`)
Sets names used in the figure for axis titles (keys ``x`` and ``y``),
colorbar title and hoverlabel (key ``color``). The values should correspond
to the desired label to be displayed. If ``img`` is an xarray, dimension
names are used for axis titles, and long name for the colorbar title
(unless overridden in ``labels``). Possible keys are: x, y, and color.
x, y: list-like, optional
x and y are used to label the axes of single-channel heatmap visualizations and
their lengths must match the lengths of the second and first dimensions of the
img argument. They are auto-populated if the input is an xarray.
animation_frame: int or str, optional (default None)
axis number along which the image array is sliced to create an animation plot.
If `img` is an xarray, `animation_frame` can be the name of one the dimensions.
facet_col: int or str, optional (default None)
axis number along which the image array is sliced to create a facetted plot.
If `img` is an xarray, `facet_col` can be the name of one the dimensions.
facet_col_wrap: int
Maximum number of facet columns. Wraps the column variable at this width,
so that the column facets span multiple rows.
Ignored if `facet_col` is None.
facet_col_spacing: float between 0 and 1
Spacing between facet columns, in paper units. Default is 0.02.
facet_row_spacing: float between 0 and 1
Spacing between facet rows created when ``facet_col_wrap`` is used, in
paper units. Default is 0.0.7.
color_continuous_scale : str or list of str
colormap used to map scalar data to colors (for a 2D image). This parameter is
not used for RGB or RGBA images. If a string is provided, it should be the name
of a known color scale, and if a list is provided, it should be a list of CSS-
compatible colors.
color_continuous_midpoint : number
If set, computes the bounds of the continuous color scale to have the desired
midpoint. Overridden by range_color or zmin and zmax.
range_color : list of two numbers
If provided, overrides auto-scaling on the continuous color scale, including
overriding `color_continuous_midpoint`. Also overrides zmin and zmax. Used only
for single-channel images.
title : str
The figure title.
template : str or dict or plotly.graph_objects.layout.Template instance
The figure template name or definition.
width : number
The figure width in pixels.
height: number
The figure height in pixels.
aspect: 'equal', 'auto', or None
- 'equal': Ensures an aspect ratio of 1 or pixels (square pixels)
- 'auto': The axes is kept fixed and the aspect ratio of pixels is
adjusted so that the data fit in the axes. In general, this will
result in non-square pixels.
- if None, 'equal' is used for numpy arrays and 'auto' for xarrays
(which have typically heterogeneous coordinates)
contrast_rescaling: 'minmax', 'infer', or None
how to determine data values corresponding to the bounds of the color
range, when zmin or zmax are not passed. If `minmax`, the min and max
values of the image are used. If `infer`, a heuristic based on the image
data type is used.
binary_string: bool, default None
if True, the image data are first rescaled and encoded as uint8 and
then passed to plotly.js as a b64 PNG string. If False, data are passed
unchanged as a numerical array. Setting to True may lead to performance
gains, at the cost of a loss of precision depending on the original data
type. If None, use_binary_string is set to True for multichannel (eg) RGB
arrays, and to False for single-channel (2D) arrays. 2D arrays are
represented as grayscale and with no colorbar if use_binary_string is
True.
binary_backend: str, 'auto' (default), 'pil' or 'pypng'
Third-party package for the transformation of numpy arrays to
png b64 strings. If 'auto', Pillow is used if installed, otherwise
pypng.
binary_compression_level: int, between 0 and 9 (default 4)
png compression level to be passed to the backend when transforming an
array to a png b64 string. Increasing `binary_compression` decreases the
size of the png string, but the compression step takes more time. For most
images it is not worth using levels greater than 5, but it's possible to
test `len(fig.data[0].source)` and to time the execution of `imshow` to
tune the level of compression. 0 means no compression (not recommended).
binary_format: str, 'png' (default) or 'jpg'
compression format used to generate b64 string. 'png' is recommended
since it uses lossless compression, but 'jpg' (lossy) compression can
result if smaller binary strings for natural images.
text_auto: bool or str (default `False`)
If `True` or a string, single-channel `img` values will be displayed as text.
A string like `'.2f'` will be interpreted as a `texttemplate` numeric formatting directive.
Returns
-------
fig : graph_objects.Figure containing the displayed image
See also
--------
plotly.graph_objects.Image : image trace
plotly.graph_objects.Heatmap : heatmap trace
Notes
-----
In order to update and customize the returned figure, use
`go.Figure.update_traces` or `go.Figure.update_layout`.
If an xarray is passed, dimensions names and coordinates are used for
axes labels and ticks.
"""
args = locals()
apply_default_cascade(args)
labels = labels.copy()
nslices_facet = 1
if facet_col is not None:
if isinstance(facet_col, str):
facet_col = img.dims.index(facet_col)
nslices_facet = img.shape[facet_col]
facet_slices = range(nslices_facet)
ncols = int(facet_col_wrap) if facet_col_wrap is not None else nslices_facet
nrows = (
nslices_facet // ncols + 1
if nslices_facet % ncols
else nslices_facet // ncols
)
else:
nrows = 1
ncols = 1
if animation_frame is not None:
if isinstance(animation_frame, str):
animation_frame = img.dims.index(animation_frame)
nslices_animation = img.shape[animation_frame]
animation_slices = range(nslices_animation)
slice_dimensions = (facet_col is not None) + (
animation_frame is not None
) # 0, 1, or 2
facet_label = None
animation_label = None
img_is_xarray = False
# ----- Define x and y, set labels if img is an xarray -------------------
if xarray_imported and isinstance(img, xarray.DataArray):
dims = list(img.dims)
img_is_xarray = True
if facet_col is not None:
facet_slices = img.coords[img.dims[facet_col]].values
_ = dims.pop(facet_col)
facet_label = img.dims[facet_col]
if animation_frame is not None:
animation_slices = img.coords[img.dims[animation_frame]].values
_ = dims.pop(animation_frame)
animation_label = img.dims[animation_frame]
y_label, x_label = dims[0], dims[1]
# np.datetime64 is not handled correctly by go.Heatmap
for ax in [x_label, y_label]:
if np.issubdtype(img.coords[ax].dtype, np.datetime64):
img.coords[ax] = img.coords[ax].astype(str)
if x is None:
x = img.coords[x_label].values
if y is None:
y = img.coords[y_label].values
if aspect is None:
aspect = "auto"
if labels.get("x", None) is None:
labels["x"] = x_label
if labels.get("y", None) is None:
labels["y"] = y_label
if labels.get("animation_frame", None) is None:
labels["animation_frame"] = animation_label
if labels.get("facet_col", None) is None:
labels["facet_col"] = facet_label
if labels.get("color", None) is None:
labels["color"] = xarray.plot.utils.label_from_attrs(img)
labels["color"] = labels["color"].replace("\n", "<br>")
else:
if hasattr(img, "columns") and hasattr(img.columns, "__len__"):
if x is None:
x = img.columns
if labels.get("x", None) is None and hasattr(img.columns, "name"):
labels["x"] = img.columns.name or ""
if hasattr(img, "index") and hasattr(img.index, "__len__"):
if y is None:
y = img.index
if labels.get("y", None) is None and hasattr(img.index, "name"):
labels["y"] = img.index.name or ""
if labels.get("x", None) is None:
labels["x"] = ""
if labels.get("y", None) is None:
labels["y"] = ""
if labels.get("color", None) is None:
labels["color"] = ""
if aspect is None:
aspect = "equal"
# --- Set the value of binary_string (forbidden for pandas)
if isinstance(img, pd.DataFrame):
if binary_string:
raise ValueError("Binary strings cannot be used with pandas arrays")
is_dataframe = True
else:
is_dataframe = False
# --------------- Starting from here img is always a numpy array --------
img = np.asanyarray(img)
# Reshape array so that animation dimension comes first, then facets, then images
if facet_col is not None:
img = np.moveaxis(img, facet_col, 0)
if animation_frame is not None and animation_frame < facet_col:
animation_frame += 1
facet_col = True
if animation_frame is not None:
img = np.moveaxis(img, animation_frame, 0)
animation_frame = True
args["animation_frame"] = (
"animation_frame"
if labels.get("animation_frame") is None
else labels["animation_frame"]
)
iterables = ()
if animation_frame is not None:
iterables += (range(nslices_animation),)
if facet_col is not None:
iterables += (range(nslices_facet),)
# Default behaviour of binary_string: True for RGB images, False for 2D
if binary_string is None:
binary_string = img.ndim >= (3 + slice_dimensions) and not is_dataframe
# Cast bools to uint8 (also one byte)
if img.dtype == np.bool:
img = 255 * img.astype(np.uint8)
if range_color is not None:
zmin = range_color[0]
zmax = range_color[1]
# -------- Contrast rescaling: either minmax or infer ------------------
if contrast_rescaling is None:
contrast_rescaling = "minmax" if img.ndim == (2 + slice_dimensions) else "infer"
# We try to set zmin and zmax only if necessary, because traces have good defaults
if contrast_rescaling == "minmax":
# When using binary_string and minmax we need to set zmin and zmax to rescale the image
if (zmin is not None or binary_string) and zmax is None:
zmax = img.max()
if (zmax is not None or binary_string) and zmin is None:
zmin = img.min()
else:
# For uint8 data and infer we let zmin and zmax to be None if passed as None
if zmax is None and img.dtype != np.uint8:
zmax = _infer_zmax_from_type(img)
if zmin is None and zmax is not None:
zmin = 0
# For 2d data, use Heatmap trace, unless binary_string is True
if img.ndim == 2 + slice_dimensions and not binary_string:
y_index = slice_dimensions
if y is not None and img.shape[y_index] != len(y):
raise ValueError(
"The length of the y vector must match the length of the first "
+ "dimension of the img matrix."
)
x_index = slice_dimensions + 1
if x is not None and img.shape[x_index] != len(x):
raise ValueError(
"The length of the x vector must match the length of the second "
+ "dimension of the img matrix."
)
texttemplate = None
if text_auto is True:
texttemplate = "%{z}"
elif text_auto is not False:
texttemplate = "%{z:" + text_auto + "}"
traces = [
go.Heatmap(
x=x,
y=y,
z=img[index_tup],
coloraxis="coloraxis1",
name=str(i),
texttemplate=texttemplate,
)
for i, index_tup in enumerate(itertools.product(*iterables))
]
autorange = True if origin == "lower" else "reversed"
layout = dict(yaxis=dict(autorange=autorange))
if aspect == "equal":
layout["xaxis"] = dict(scaleanchor="y", constrain="domain")
layout["yaxis"]["constrain"] = "domain"
colorscale_validator = ColorscaleValidator("colorscale", "imshow")
layout["coloraxis1"] = dict(
colorscale=colorscale_validator.validate_coerce(
args["color_continuous_scale"]
),
cmid=color_continuous_midpoint,
cmin=zmin,
cmax=zmax,
)
if labels["color"]:
layout["coloraxis1"]["colorbar"] = dict(title_text=labels["color"])
# For 2D+RGB data, use Image trace
elif (
img.ndim >= 3
and (img.shape[-1] in [3, 4] or slice_dimensions and binary_string)
) or (img.ndim == 2 and binary_string):
rescale_image = True # to check whether image has been modified
if zmin is not None and zmax is not None:
zmin, zmax = (
_vectorize_zvalue(zmin, mode="min"),
_vectorize_zvalue(zmax, mode="max"),
)
x0, y0, dx, dy = (None,) * 4
error_msg_xarray = (
"Non-numerical coordinates were passed with xarray `img`, but "
"the Image trace cannot handle it. Please use `binary_string=False` "
"for 2D data or pass instead the numpy array `img.values` to `px.imshow`."
)
if x is not None:
x = np.asanyarray(x)
if np.issubdtype(x.dtype, np.number):
x0 = x[0]
dx = x[1] - x[0]
else:
error_msg = (
error_msg_xarray
if img_is_xarray
else (
"Only numerical values are accepted for the `x` parameter "
"when an Image trace is used."
)
)
raise ValueError(error_msg)
if y is not None:
y = np.asanyarray(y)
if np.issubdtype(y.dtype, np.number):
y0 = y[0]
dy = y[1] - y[0]
else:
error_msg = (
error_msg_xarray
if img_is_xarray
else (
"Only numerical values are accepted for the `y` parameter "
"when an Image trace is used."
)
)
raise ValueError(error_msg)
if binary_string:
if zmin is None and zmax is None: # no rescaling, faster
img_rescaled = img
rescale_image = False
elif img.ndim == 2 + slice_dimensions: # single-channel image
img_rescaled = rescale_intensity(
img, in_range=(zmin[0], zmax[0]), out_range=np.uint8
)
else:
img_rescaled = np.stack(
[
rescale_intensity(
img[..., ch],
in_range=(zmin[ch], zmax[ch]),
out_range=np.uint8,
)
for ch in range(img.shape[-1])
],
axis=-1,
)
img_str = [
image_array_to_data_uri(
img_rescaled[index_tup],
backend=binary_backend,
compression=binary_compression_level,
ext=binary_format,
)
for index_tup in itertools.product(*iterables)
]
traces = [
go.Image(source=img_str_slice, name=str(i), x0=x0, y0=y0, dx=dx, dy=dy)
for i, img_str_slice in enumerate(img_str)
]
else:
colormodel = "rgb" if img.shape[-1] == 3 else "rgba256"
traces = [
go.Image(
z=img[index_tup],
zmin=zmin,
zmax=zmax,
colormodel=colormodel,
x0=x0,
y0=y0,
dx=dx,
dy=dy,
)
for index_tup in itertools.product(*iterables)
]
layout = {}
if origin == "lower" or (dy is not None and dy < 0):
layout["yaxis"] = dict(autorange=True)
if dx is not None and dx < 0:
layout["xaxis"] = dict(autorange="reversed")
else:
raise ValueError(
"px.imshow only accepts 2D single-channel, RGB or RGBA images. "
"An image of shape %s was provided. "
"Alternatively, 3- or 4-D single or multichannel datasets can be "
"visualized using the `facet_col` or/and `animation_frame` arguments."
% str(img.shape)
)
# Now build figure
col_labels = []
if facet_col is not None:
slice_label = (
"facet_col" if labels.get("facet_col") is None else labels["facet_col"]
)
col_labels = ["%s=%d" % (slice_label, i) for i in facet_slices]
fig = init_figure(args, "xy", [], nrows, ncols, col_labels, [])
for attr_name in ["height", "width"]:
if args[attr_name]:
layout[attr_name] = args[attr_name]
if args["title"]:
layout["title_text"] = args["title"]
elif args["template"].layout.margin.t is None:
layout["margin"] = {"t": 60}
frame_list = []
for index, trace in enumerate(traces):
if (facet_col and index < nrows * ncols) or index == 0:
fig.add_trace(trace, row=nrows - index // ncols, col=index % ncols + 1)
if animation_frame is not None:
for i, index in zip(range(nslices_animation), animation_slices):
frame_list.append(
dict(
data=traces[nslices_facet * i : nslices_facet * (i + 1)],
layout=layout,
name=str(index),
)
)
if animation_frame:
fig.frames = frame_list
fig.update_layout(layout)
# Hover name, z or color
if binary_string and rescale_image and not np.all(img == img_rescaled):
# we rescaled the image, hence z is not displayed in hover since it does
# not correspond to img values
hovertemplate = "%s: %%{x}<br>%s: %%{y}<extra></extra>" % (
labels["x"] or "x",
labels["y"] or "y",
)
else:
if trace["type"] == "heatmap":
hover_name = "%{z}"
elif img.ndim == 2:
hover_name = "%{z[0]}"
elif img.ndim == 3 and img.shape[-1] == 3:
hover_name = "[%{z[0]}, %{z[1]}, %{z[2]}]"
else:
hover_name = "%{z}"
hovertemplate = "%s: %%{x}<br>%s: %%{y}<br>%s: %s<extra></extra>" % (
labels["x"] or "x",
labels["y"] or "y",
labels["color"] or "color",
hover_name,
)
fig.update_traces(hovertemplate=hovertemplate)
if labels["x"]:
fig.update_xaxes(title_text=labels["x"], row=1)
if labels["y"]:
fig.update_yaxes(title_text=labels["y"], col=1)
configure_animation_controls(args, go.Image, fig)
fig.update_layout(template=args["template"], overwrite=True)
return fig
# Now we'll be building a new classifier that takes in our initial predictions as a pre-train model
base_predictions_train = pd.DataFrame({
'RandomForest': rf_oof_train.ravel(),
'ExtraTrees': et_oof_train.ravel(),
'AdaBoost': ada_oof_train.ravel(),
'GradientBoost': gb_oof_train.ravel()
})
base_predictions_train.head()
# In[ ]:
# correlation heatmap of second-level training
data = [
go.Heatmap(x=base_predictions_train.columns.values,
y=base_predictions_train.columns.values,
colorscale='Viridis',
showscale=True,
reversescale=True,
z=base_predictions_train.astype(float).corr().values)
]
py.iplot(data, filename='labeled-headmap')
# In[ ]:
x_train = np.concatenate(
(et_oof_train, rf_oof_train, ada_oof_train, gb_oof_train, svc_oof_train),
axis=1)
x_test = np.concatenate(
(et_oof_test, rf_oof_test, ada_oof_test, gb_oof_test, svc_oof_test),
axis=1)
# In[ ]:
def plot_decision_boundary_and_metrics(
model, x_train, y_train, x_test, y_test, metrics
):
d = x_train.shape[1]
x_min, x_max = x_train[:, 0].min() - 1, x_train[:, 0].max() + 1
y_min, y_max = x_train[:, 1].min() - 1, x_train[:, 1].max() + 1
h = 0.02
xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))
y_ = np.arange(y_min, y_max, h)
model_input = [(xx.ravel() ** p, yy.ravel() ** p) for p in range(1, d // 2 + 1)]
aux = []
for c in model_input:
aux.append(c[0])
aux.append(c[1])
Z = model.predict(np.concatenate([v.reshape(-1, 1) for v in aux], axis=1))
Z = Z.reshape(xx.shape)
fig = make_subplots(
rows=2,
cols=2,
specs=[[{"colspan": 2}, None], [{"type": "indicator"}, {"type": "indicator"}]],
subplot_titles=("Decision Boundary", None, None),
row_heights=[0.7, 0.30],
)
heatmap = go.Heatmap(
x=xx[0],
y=y_,
z=Z,
colorscale=["tomato", "rgb(27,158,119)"],
showscale=False,
)
train_data = go.Scatter(
x=x_train[:, 0],
y=x_train[:, 1],
name="train data",
mode="markers",
showlegend=True,
marker=dict(
size=10,
color=y_train,
colorscale=["tomato", "green"],
line=dict(color="black", width=2),
),
)
test_data = go.Scatter(
x=x_test[:, 0],
y=x_test[:, 1],
name="test data",
mode="markers",
showlegend=True,
marker_symbol="cross",
visible="legendonly",
marker=dict(
size=10,
color=y_test,
colorscale=["tomato", "green"],
line=dict(color="black", width=2),
),
)
fig.add_trace(heatmap, row=1, col=1,).add_trace(train_data).add_trace(
test_data
).update_xaxes(range=[x_min, x_max], title="x1").update_yaxes(
range=[y_min, y_max], title="x2"
)
fig.add_trace(
go.Indicator(
mode="gauge+number+delta",
value=metrics["test_accuracy"],
title={"text": f"Accuracy (test)"},
domain={"x": [0, 1], "y": [0, 1]},
gauge={"axis": {"range": [0, 1]}},
delta={"reference": metrics["train_accuracy"]},
),
row=2,
col=1,
)
fig.add_trace(
go.Indicator(
mode="gauge+number+delta",
value=metrics["test_f1"],
title={"text": f"F1 score (test)"},
domain={"x": [0, 1], "y": [0, 1]},
gauge={"axis": {"range": [0, 1]}},
delta={"reference": metrics["train_f1"]},
),
row=2,
col=2,
)
fig.update_layout(
height=700,
)
return fig
y='value',
color='query',
hover_name="query").for_each_trace(
lambda t: t.update(name=t.name.split("=")[1]))
fig.update_traces(mode='markers+lines')
fig.update_layout(legend_traceorder="reversed")
st.plotly_chart(fig)
if selection == 'Карта корреляций':
cov_corr_t = pd.read_excel('data/corr_true_data.xlsx')
cov_pivot = pd.pivot_table(cov_corr_t,
values='corr',
index=['A'],
columns=['B'])
fig = go.Figure(data=go.Heatmap(z=cov_pivot,
x=cov_pivot.columns,
y=cov_pivot.columns,
hoverongaps=False,
colorscale='deep'))
fig.show()
if selection == 'Поиск взаимосвязей':
st.header('Поиск взаимосвязей. Методика расчета')
st.markdown(
"""Для поиска связей в данных были подготовленны датасеты помесячных данных релевантной поисковой статистики и статистики COVID-19<br>
Далее, был проведен поиск корреляций Пирсона между всеми наборами данных данного датасета<br>
И к полученным результатам применена поправка Бенджамини-Хохберга на множественную проверку гипотез, для повышения надежности результатов. <br>
Подтвержденные значения корреляций показаны ниже.
""",
unsafe_allow_html=True)
df_cov_true = pd.read_excel('data/corr_true_data.xlsx')
st.write(df_cov_true)
st.markdown("""corr - значение корреляции пирсона между A и B<br>
def update_figure(selected_config, selected_has, selected_im, selected_nant,
selected_zoom, selected_declination, selected_latitude,
selected_weighting):
ctx = dash.callback_context
# print 'Trig : ', ctx.triggered
# print 'States : ', ctx.states
# print 'Inputs : ', ctx.inputs
#Trig : [{'prop_id': u'config-dropdown.value', 'value': u'RandomConfig'}]
thischanged = None
if ctx.triggered[0]['prop_id'].count('config-dropdown'):
thischanged = 'config'
if ctx.triggered[0]['prop_id'].count('source-dropdown'):
thischanged = 'source'
if ctx.triggered[0]['prop_id'].count('timerange-slider'):
thischanged = 'timerange'
if ctx.triggered[0]['prop_id'].count('zoom-slider'):
thischanged = 'antzoom'
if ctx.triggered[0]['prop_id'].count('nant-dropdown'):
thischanged = 'nant'
if ctx.triggered[0]['prop_id'].count('declination-picker'):
thischanged = 'declination'
if ctx.triggered[0]['prop_id'].count('latitude-picker'):
thischanged = 'latitude'
if ctx.triggered[0]['prop_id'].count('weighting-picker'):
thischanged = 'weighting'
#print 'This changed : ', thischanged
tim1 = time.time()
if thischanged == 'config' or thischanged == 'antzoom' or thischanged == 'nant':
changeseed = True
if thischanged == 'antzoom' or thischanged == 'nant':
changeseed = False
if eval(selected_nant) == 2:
selected_config = 'RandomConfig'
print("Using random locations for 2 antennas")
tel.calcAntList(configtype=selected_config,
zoom=2**selected_zoom,
changeseed=changeseed,
nant=eval(selected_nant))
# if thischanged=='antzoom':
# tel.scaleAntList(zoom=selected_zoom)
antlist = tel.getAntList()
antlocsX = antlist['EastLoc']
antlocsY = antlist['NorthLoc']
tim2 = time.time()
# print antlocsX, antlocsY
### Prepare contents of antenna layout plot
traces1 = []
traces1.append(
go.Scatter(
x=antlocsX,
y=antlocsY,
text=selected_config,
mode='markers',
opacity=0.7,
marker={
'size': 15,
'line': {
'width': 0.5,
'color': 'white'
}
},
))
tim3 = time.time()
### Prepare contents of observed image.
### Slider callback too
if thischanged == 'config' or thischanged == 'timerange' or thischanged == 'antzoom' or thischanged == 'nant' or thischanged == 'declination' or thischanged == 'latitude' or thischanged == 'weighting':
tel.makeUVcov(has=selected_has,
dec=selected_declination,
obslatitude=selected_latitude,
weighting=selected_weighting)
tim4 = time.time()
traces3 = []
traces3.append(go.Heatmap(z=tel.getUVcov()))
## sky callback
if thischanged == 'source':
tel.setsky(imtype=selected_im)
tim5 = time.time()
## Sim
tel.makeImage()
tim6 = time.time()
traces2 = []
traces2.append(go.Heatmap(z=tel.getImage()))
tim7 = time.time()
# print 'Ant read time : ', tim2-tim1
# print 'Ant plot time : ', tim3-tim2
# print 'UVcov time : ', tim4-tim3
# print 'Source time : ', tim5-tim4
# print 'Observe time :', tim6-tim5
# print 'Make raster : ', tim7-tim6
# print 'Total time in callback : ', tim7-tim1, ' sec'
### Return list of outputs, sync'd with specification above
dispsize = 400
return [{
'data':
traces1,
'layout':
go.Layout(xaxis={
'title': 'X Position (m)',
'range': [-1000, 1000]
},
yaxis={
'title': 'Y Position (m)',
'range': [-1000, 1000]
},
title="ARRAY CONFIGURATION",
width=dispsize,
height=dispsize,
autosize=True)
}, {
'data':
traces3,
'layout':
go.Layout(xaxis={'title': 'Spatial frequency : U (pixels)'},
yaxis={'title': 'Spatial frequency : V (pixels)'},
title="SPATIAL FREQUENCY COVERAGE",
width=dispsize,
height=dispsize,
autosize=True)
}, {
'data':
traces2,
'layout':
go.Layout(xaxis={'title': 'Right Ascension (pixels)'},
yaxis={'title': 'Declination (pixels)'},
title="OBSERVED IMAGE",
width=dispsize,
height=dispsize,
autosize=True)
}]
def plot_covariance_returns_correlation(correlation, title):
config = helper.generate_config()
graph_path = "graphs/{}.html".format(_sanatize_string(title))
data = []
dendro_top = ff.create_dendrogram(correlation, orientation="bottom")
for i in range(len(dendro_top["data"])):
dendro_top["data"][i]["yaxis"] = "y2"
data.extend(dendro_top["data"])
dendro_left = ff.create_dendrogram(correlation, orientation="right")
for i in range(len(dendro_left["data"])):
dendro_left["data"][i]["xaxis"] = "x2"
data.extend(dendro_left["data"])
heatmap_hover_text = _generate_hover_text(
correlation.index,
correlation.columns,
correlation.values,
"Ticker 2",
"Ticker 1",
"Correlation",
)
heatmap_trace = go.Heatmap(
x=dendro_top["layout"]["xaxis"]["tickvals"],
y=dendro_left["layout"]["yaxis"]["tickvals"],
z=correlation.values,
zauto=False,
zmax=1.0,
zmin=-1.0,
text=heatmap_hover_text,
hoverinfo="text",
)
data.append(heatmap_trace)
xaxis1_layout = {
"showgrid": False,
"showline": False,
"zeroline": False,
"showticklabels": False,
"ticks": "",
}
xaxis2_layout = {
"showgrid": False,
"zeroline": False,
"showticklabels": False
}
layout = go.Layout(title=title, showlegend=False, width=800, height=800)
figure = go.Figure(data=data, layout=layout)
figure["layout"]["xaxis"].update({"domain": [0.15, 1]})
figure["layout"]["xaxis"].update(xaxis1_layout)
figure["layout"]["yaxis"].update({"domain": [0, 0.85]})
figure["layout"]["yaxis"].update(xaxis1_layout)
figure["layout"]["xaxis2"].update({"domain": [0, 0.15]})
figure["layout"]["xaxis2"].update(xaxis2_layout)
figure["layout"]["yaxis2"].update({"domain": [0.825, 0.975]})
figure["layout"]["yaxis2"].update(xaxis2_layout)
offline_py.plot(figure,
config=config,
filename=graph_path,
auto_open=False)
display(
HTML(
'The graph for {} is too large. You can view it <a href="{}" target="_blank">here</a>.'
.format(title, graph_path)))
def imshow(
img,
zmin=None,
zmax=None,
origin=None,
labels={},
x=None,
y=None,
color_continuous_scale=None,
color_continuous_midpoint=None,
range_color=None,
title=None,
template=None,
width=None,
height=None,
aspect=None,
):
"""
Display an image, i.e. data on a 2D regular raster.
Parameters
----------
img: array-like image, or xarray
The image data. Supported array shapes are
- (M, N): an image with scalar data. The data is visualized
using a colormap.
- (M, N, 3): an image with RGB values.
- (M, N, 4): an image with RGBA values, i.e. including transparency.
zmin, zmax : scalar or iterable, optional
zmin and zmax define the scalar range that the colormap covers. By default,
zmin and zmax correspond to the min and max values of the datatype for integer
datatypes (ie [0-255] for uint8 images, [0, 65535] for uint16 images, etc.). For
a multichannel image of floats, the max of the image is computed and zmax is the
smallest power of 256 (1, 255, 65535) greater than this max value,
with a 5% tolerance. For a single-channel image, the max of the image is used.
Overridden by range_color.
origin : str, 'upper' or 'lower' (default 'upper')
position of the [0, 0] pixel of the image array, in the upper left or lower left
corner. The convention 'upper' is typically used for matrices and images.
labels : dict with str keys and str values (default `{}`)
Sets names used in the figure for axis titles (keys ``x`` and ``y``),
colorbar title and hoverlabel (key ``color``). The values should correspond
to the desired label to be displayed. If ``img`` is an xarray, dimension
names are used for axis titles, and long name for the colorbar title
(unless overridden in ``labels``). Possible keys are: x, y, and color.
x, y: list-like, optional
x and y are used to label the axes of single-channel heatmap visualizations and
their lengths must match the lengths of the second and first dimensions of the
img argument. They are auto-populated if the input is an xarray.
color_continuous_scale : str or list of str
colormap used to map scalar data to colors (for a 2D image). This parameter is
not used for RGB or RGBA images. If a string is provided, it should be the name
of a known color scale, and if a list is provided, it should be a list of CSS-
compatible colors.
color_continuous_midpoint : number
If set, computes the bounds of the continuous color scale to have the desired
midpoint. Overridden by range_color or zmin and zmax.
range_color : list of two numbers
If provided, overrides auto-scaling on the continuous color scale, including
overriding `color_continuous_midpoint`. Also overrides zmin and zmax. Used only
for single-channel images.
title : str
The figure title.
template : str or dict or plotly.graph_objects.layout.Template instance
The figure template name or definition.
width : number
The figure width in pixels.
height: number
The figure height in pixels.
aspect: 'equal', 'auto', or None
- 'equal': Ensures an aspect ratio of 1 or pixels (square pixels)
- 'auto': The axes is kept fixed and the aspect ratio of pixels is
adjusted so that the data fit in the axes. In general, this will
result in non-square pixels.
- if None, 'equal' is used for numpy arrays and 'auto' for xarrays
(which have typically heterogeneous coordinates)
Returns
-------
fig : graph_objects.Figure containing the displayed image
See also
--------
plotly.graph_objects.Image : image trace
plotly.graph_objects.Heatmap : heatmap trace
Notes
-----
In order to update and customize the returned figure, use
`go.Figure.update_traces` or `go.Figure.update_layout`.
If an xarray is passed, dimensions names and coordinates are used for
axes labels and ticks.
"""
args = locals()
apply_default_cascade(args)
labels = labels.copy()
if xarray_imported and isinstance(img, xarray.DataArray):
y_label, x_label = img.dims[0], img.dims[1]
# np.datetime64 is not handled correctly by go.Heatmap
for ax in [x_label, y_label]:
if np.issubdtype(img.coords[ax].dtype, np.datetime64):
img.coords[ax] = img.coords[ax].astype(str)
if x is None:
x = img.coords[x_label]
if y is None:
y = img.coords[y_label]
if aspect is None:
aspect = "auto"
if labels.get("x", None) is None:
labels["x"] = x_label
if labels.get("y", None) is None:
labels["y"] = y_label
if labels.get("color", None) is None:
labels["color"] = xarray.plot.utils.label_from_attrs(img)
labels["color"] = labels["color"].replace("\n", "<br>")
else:
if labels.get("x", None) is None:
labels["x"] = ""
if labels.get("y", None) is None:
labels["y"] = ""
if labels.get("color", None) is None:
labels["color"] = ""
if aspect is None:
aspect = "equal"
img = np.asanyarray(img)
# Cast bools to uint8 (also one byte)
if img.dtype == np.bool:
img = 255 * img.astype(np.uint8)
# For 2d data, use Heatmap trace
if img.ndim == 2:
if y is not None and img.shape[0] != len(y):
raise ValueError(
"The length of the y vector must match the length of the first "
+ "dimension of the img matrix.")
if x is not None and img.shape[1] != len(x):
raise ValueError(
"The length of the x vector must match the length of the second "
+ "dimension of the img matrix.")
trace = go.Heatmap(x=x, y=y, z=img, coloraxis="coloraxis1")
autorange = True if origin == "lower" else "reversed"
layout = dict(yaxis=dict(autorange=autorange))
if aspect == "equal":
layout["xaxis"] = dict(scaleanchor="y", constrain="domain")
layout["yaxis"]["constrain"] = "domain"
colorscale_validator = ColorscaleValidator("colorscale", "imshow")
if zmin is not None and zmax is None:
zmax = img.max()
if zmax is not None and zmin is None:
zmin = img.min()
range_color = range_color or [zmin, zmax]
layout["coloraxis1"] = dict(
colorscale=colorscale_validator.validate_coerce(
args["color_continuous_scale"]),
cmid=color_continuous_midpoint,
cmin=range_color[0],
cmax=range_color[1],
)
if labels["color"]:
layout["coloraxis1"]["colorbar"] = dict(title_text=labels["color"])
# For 2D+RGB data, use Image trace
elif img.ndim == 3 and img.shape[-1] in [3, 4]:
if zmax is None and img.dtype is not np.uint8:
zmax = _infer_zmax_from_type(img)
zmin, zmax = _vectorize_zvalue(zmin), _vectorize_zvalue(zmax)
trace = go.Image(z=img, zmin=zmin, zmax=zmax)
layout = {}
if origin == "lower":
layout["yaxis"] = dict(autorange=True)
else:
raise ValueError(
"px.imshow only accepts 2D single-channel, RGB or RGBA images. "
"An image of shape %s was provided" % str(img.shape))
layout_patch = dict()
for attr_name in ["height", "width"]:
if args[attr_name]:
layout_patch[attr_name] = args[attr_name]
if args["title"]:
layout_patch["title_text"] = args["title"]
elif args["template"].layout.margin.t is None:
layout_patch["margin"] = {"t": 60}
fig = go.Figure(data=trace, layout=layout)
fig.update_layout(layout_patch)
fig.update_traces(
hovertemplate="%s: %%{x}<br>%s: %%{y}<br>%s: %%{z}<extra></extra>" % (
labels["x"] or "x",
labels["y"] or "y",
labels["color"] or "color",
))
if labels["x"]:
fig.update_xaxes(title_text=labels["x"])
if labels["y"]:
fig.update_yaxes(title_text=labels["y"])
fig.update_layout(template=args["template"], overwrite=True)
return fig
def imshow(
img,
zmin=None,
zmax=None,
origin=None,
labels={},
x=None,
y=None,
color_continuous_scale=None,
color_continuous_midpoint=None,
range_color=None,
title=None,
template=None,
width=None,
height=None,
aspect=None,
contrast_rescaling=None,
binary_string=None,
binary_backend="auto",
binary_compression_level=4,
binary_format="png",
):
"""
Display an image, i.e. data on a 2D regular raster.
Parameters
----------
img: array-like image, or xarray
The image data. Supported array shapes are
- (M, N): an image with scalar data. The data is visualized
using a colormap.
- (M, N, 3): an image with RGB values.
- (M, N, 4): an image with RGBA values, i.e. including transparency.
zmin, zmax : scalar or iterable, optional
zmin and zmax define the scalar range that the colormap covers. By default,
zmin and zmax correspond to the min and max values of the datatype for integer
datatypes (ie [0-255] for uint8 images, [0, 65535] for uint16 images, etc.). For
a multichannel image of floats, the max of the image is computed and zmax is the
smallest power of 256 (1, 255, 65535) greater than this max value,
with a 5% tolerance. For a single-channel image, the max of the image is used.
Overridden by range_color.
origin : str, 'upper' or 'lower' (default 'upper')
position of the [0, 0] pixel of the image array, in the upper left or lower left
corner. The convention 'upper' is typically used for matrices and images.
labels : dict with str keys and str values (default `{}`)
Sets names used in the figure for axis titles (keys ``x`` and ``y``),
colorbar title and hoverlabel (key ``color``). The values should correspond
to the desired label to be displayed. If ``img`` is an xarray, dimension
names are used for axis titles, and long name for the colorbar title
(unless overridden in ``labels``). Possible keys are: x, y, and color.
x, y: list-like, optional
x and y are used to label the axes of single-channel heatmap visualizations and
their lengths must match the lengths of the second and first dimensions of the
img argument. They are auto-populated if the input is an xarray.
color_continuous_scale : str or list of str
colormap used to map scalar data to colors (for a 2D image). This parameter is
not used for RGB or RGBA images. If a string is provided, it should be the name
of a known color scale, and if a list is provided, it should be a list of CSS-
compatible colors.
color_continuous_midpoint : number
If set, computes the bounds of the continuous color scale to have the desired
midpoint. Overridden by range_color or zmin and zmax.
range_color : list of two numbers
If provided, overrides auto-scaling on the continuous color scale, including
overriding `color_continuous_midpoint`. Also overrides zmin and zmax. Used only
for single-channel images.
title : str
The figure title.
template : str or dict or plotly.graph_objects.layout.Template instance
The figure template name or definition.
width : number
The figure width in pixels.
height: number
The figure height in pixels.
aspect: 'equal', 'auto', or None
- 'equal': Ensures an aspect ratio of 1 or pixels (square pixels)
- 'auto': The axes is kept fixed and the aspect ratio of pixels is
adjusted so that the data fit in the axes. In general, this will
result in non-square pixels.
- if None, 'equal' is used for numpy arrays and 'auto' for xarrays
(which have typically heterogeneous coordinates)
contrast_rescaling: 'minmax', 'infer', or None
how to determine data values corresponding to the bounds of the color
range, when zmin or zmax are not passed. If `minmax`, the min and max
values of the image are used. If `infer`, a heuristic based on the image
data type is used.
binary_string: bool, default None
if True, the image data are first rescaled and encoded as uint8 and
then passed to plotly.js as a b64 PNG string. If False, data are passed
unchanged as a numerical array. Setting to True may lead to performance
gains, at the cost of a loss of precision depending on the original data
type. If None, use_binary_string is set to True for multichannel (eg) RGB
arrays, and to False for single-channel (2D) arrays. 2D arrays are
represented as grayscale and with no colorbar if use_binary_string is
True.
binary_backend: str, 'auto' (default), 'pil' or 'pypng'
Third-party package for the transformation of numpy arrays to
png b64 strings. If 'auto', Pillow is used if installed, otherwise
pypng.
binary_compression_level: int, between 0 and 9 (default 4)
png compression level to be passed to the backend when transforming an
array to a png b64 string. Increasing `binary_compression` decreases the
size of the png string, but the compression step takes more time. For most
images it is not worth using levels greater than 5, but it's possible to
test `len(fig.data[0].source)` and to time the execution of `imshow` to
tune the level of compression. 0 means no compression (not recommended).
binary_format: str, 'png' (default) or 'jpg'
compression format used to generate b64 string. 'png' is recommended
since it uses lossless compression, but 'jpg' (lossy) compression can
result if smaller binary strings for natural images.
Returns
-------
fig : graph_objects.Figure containing the displayed image
See also
--------
plotly.graph_objects.Image : image trace
plotly.graph_objects.Heatmap : heatmap trace
Notes
-----
In order to update and customize the returned figure, use
`go.Figure.update_traces` or `go.Figure.update_layout`.
If an xarray is passed, dimensions names and coordinates are used for
axes labels and ticks.
"""
args = locals()
apply_default_cascade(args)
labels = labels.copy()
# ----- Define x and y, set labels if img is an xarray -------------------
if xarray_imported and isinstance(img, xarray.DataArray):
if binary_string:
raise ValueError(
"It is not possible to use binary image strings for xarrays."
"Please pass your data as a numpy array instead using"
"`img.values`"
)
y_label, x_label = img.dims[0], img.dims[1]
# np.datetime64 is not handled correctly by go.Heatmap
for ax in [x_label, y_label]:
if np.issubdtype(img.coords[ax].dtype, np.datetime64):
img.coords[ax] = img.coords[ax].astype(str)
if x is None:
x = img.coords[x_label]
if y is None:
y = img.coords[y_label]
if aspect is None:
aspect = "auto"
if labels.get("x", None) is None:
labels["x"] = x_label
if labels.get("y", None) is None:
labels["y"] = y_label
if labels.get("color", None) is None:
labels["color"] = xarray.plot.utils.label_from_attrs(img)
labels["color"] = labels["color"].replace("\n", "<br>")
else:
if hasattr(img, "columns") and hasattr(img.columns, "__len__"):
if x is None:
x = img.columns
if labels.get("x", None) is None and hasattr(img.columns, "name"):
labels["x"] = img.columns.name or ""
if hasattr(img, "index") and hasattr(img.index, "__len__"):
if y is None:
y = img.index
if labels.get("y", None) is None and hasattr(img.index, "name"):
labels["y"] = img.index.name or ""
if labels.get("x", None) is None:
labels["x"] = ""
if labels.get("y", None) is None:
labels["y"] = ""
if labels.get("color", None) is None:
labels["color"] = ""
if aspect is None:
aspect = "equal"
# --- Set the value of binary_string (forbidden for pandas)
if isinstance(img, pd.DataFrame):
if binary_string:
raise ValueError("Binary strings cannot be used with pandas arrays")
is_dataframe = True
else:
is_dataframe = False
# --------------- Starting from here img is always a numpy array --------
img = np.asanyarray(img)
# Default behaviour of binary_string: True for RGB images, False for 2D
if binary_string is None:
binary_string = img.ndim >= 3 and not is_dataframe
# Cast bools to uint8 (also one byte)
if img.dtype == np.bool:
img = 255 * img.astype(np.uint8)
if range_color is not None:
zmin = range_color[0]
zmax = range_color[1]
# -------- Contrast rescaling: either minmax or infer ------------------
if contrast_rescaling is None:
contrast_rescaling = "minmax" if img.ndim == 2 else "infer"
# We try to set zmin and zmax only if necessary, because traces have good defaults
if contrast_rescaling == "minmax":
# When using binary_string and minmax we need to set zmin and zmax to rescale the image
if (zmin is not None or binary_string) and zmax is None:
zmax = img.max()
if (zmax is not None or binary_string) and zmin is None:
zmin = img.min()
else:
# For uint8 data and infer we let zmin and zmax to be None if passed as None
if zmax is None and img.dtype != np.uint8:
zmax = _infer_zmax_from_type(img)
if zmin is None and zmax is not None:
zmin = 0
# For 2d data, use Heatmap trace, unless binary_string is True
if img.ndim == 2 and not binary_string:
if y is not None and img.shape[0] != len(y):
raise ValueError(
"The length of the y vector must match the length of the first "
+ "dimension of the img matrix."
)
if x is not None and img.shape[1] != len(x):
raise ValueError(
"The length of the x vector must match the length of the second "
+ "dimension of the img matrix."
)
trace = go.Heatmap(x=x, y=y, z=img, coloraxis="coloraxis1")
autorange = True if origin == "lower" else "reversed"
layout = dict(yaxis=dict(autorange=autorange))
if aspect == "equal":
layout["xaxis"] = dict(scaleanchor="y", constrain="domain")
layout["yaxis"]["constrain"] = "domain"
colorscale_validator = ColorscaleValidator("colorscale", "imshow")
layout["coloraxis1"] = dict(
colorscale=colorscale_validator.validate_coerce(
args["color_continuous_scale"]
),
cmid=color_continuous_midpoint,
cmin=zmin,
cmax=zmax,
)
if labels["color"]:
layout["coloraxis1"]["colorbar"] = dict(title_text=labels["color"])
# For 2D+RGB data, use Image trace
elif img.ndim == 3 and img.shape[-1] in [3, 4] or (img.ndim == 2 and binary_string):
rescale_image = True # to check whether image has been modified
if zmin is not None and zmax is not None:
zmin, zmax = (
_vectorize_zvalue(zmin, mode="min"),
_vectorize_zvalue(zmax, mode="max"),
)
if binary_string:
if zmin is None and zmax is None: # no rescaling, faster
img_rescaled = img
rescale_image = False
elif img.ndim == 2:
img_rescaled = rescale_intensity(
img, in_range=(zmin[0], zmax[0]), out_range=np.uint8
)
else:
img_rescaled = np.dstack(
[
rescale_intensity(
img[..., ch],
in_range=(zmin[ch], zmax[ch]),
out_range=np.uint8,
)
for ch in range(img.shape[-1])
]
)
img_str = _array_to_b64str(
img_rescaled,
backend=binary_backend,
compression=binary_compression_level,
ext=binary_format,
)
trace = go.Image(source=img_str)
else:
colormodel = "rgb" if img.shape[-1] == 3 else "rgba256"
trace = go.Image(z=img, zmin=zmin, zmax=zmax, colormodel=colormodel)
layout = {}
if origin == "lower":
layout["yaxis"] = dict(autorange=True)
else:
raise ValueError(
"px.imshow only accepts 2D single-channel, RGB or RGBA images. "
"An image of shape %s was provided" % str(img.shape)
)
layout_patch = dict()
for attr_name in ["height", "width"]:
if args[attr_name]:
layout_patch[attr_name] = args[attr_name]
if args["title"]:
layout_patch["title_text"] = args["title"]
elif args["template"].layout.margin.t is None:
layout_patch["margin"] = {"t": 60}
fig = go.Figure(data=trace, layout=layout)
fig.update_layout(layout_patch)
# Hover name, z or color
if binary_string and rescale_image and not np.all(img == img_rescaled):
# we rescaled the image, hence z is not displayed in hover since it does
# not correspond to img values
hovertemplate = "%s: %%{x}<br>%s: %%{y}<extra></extra>" % (
labels["x"] or "x",
labels["y"] or "y",
)
else:
if trace["type"] == "heatmap":
hover_name = "%{z}"
elif img.ndim == 2:
hover_name = "%{z[0]}"
elif img.ndim == 3 and img.shape[-1] == 3:
hover_name = "[%{z[0]}, %{z[1]}, %{z[2]}]"
else:
hover_name = "%{z}"
hovertemplate = "%s: %%{x}<br>%s: %%{y}<br>%s: %s<extra></extra>" % (
labels["x"] or "x",
labels["y"] or "y",
labels["color"] or "color",
hover_name,
)
fig.update_traces(hovertemplate=hovertemplate)
if labels["x"]:
fig.update_xaxes(title_text=labels["x"])
if labels["y"]:
fig.update_yaxes(title_text=labels["y"])
fig.update_layout(template=args["template"], overwrite=True)
return fig
def plot_nf_heatmap(plot, input_data):
"""Generate the plot(s) with algorithm: plot_nf_heatmap
specified in the specification file.
:param plot: Plot to generate.
:param input_data: Data to process.
:type plot: pandas.Series
:type input_data: InputData
"""
regex_cn = re.compile(r'^(\d*)R(\d*)C$')
regex_test_name = re.compile(r'^.*-(\d+ch|\d+pl)-'
r'(\d+mif|\d+vh)-'
r'(\d+vm\d+t|\d+dcr\d+t|\d+dcr\d+c).*$')
vals = dict()
# Transform the data
logging.info(f" Creating the data set for the {plot.get(u'type', u'')} "
f"{plot.get(u'title', u'')}.")
data = input_data.filter_data(plot, continue_on_error=True)
if data is None or data.empty:
logging.error(u"No data.")
return
for job in data:
for build in job:
for test in build:
for tag in test[u"tags"]:
groups = re.search(regex_cn, tag)
if groups:
chain = str(groups.group(1))
node = str(groups.group(2))
break
else:
continue
groups = re.search(regex_test_name, test[u"name"])
if groups and len(groups.groups()) == 3:
hover_name = (f"{str(groups.group(1))}-"
f"{str(groups.group(2))}-"
f"{str(groups.group(3))}")
else:
hover_name = u""
if vals.get(chain, None) is None:
vals[chain] = dict()
if vals[chain].get(node, None) is None:
vals[chain][node] = dict(name=hover_name,
vals=list(),
nr=None,
mean=None,
stdev=None)
try:
if plot[u"include-tests"] == u"MRR":
result = test[u"result"][u"receive-rate"]
elif plot[u"include-tests"] == u"PDR":
result = test[u"throughput"][u"PDR"][u"LOWER"]
elif plot[u"include-tests"] == u"NDR":
result = test[u"throughput"][u"NDR"][u"LOWER"]
else:
result = None
except TypeError:
result = None
if result:
vals[chain][node][u"vals"].append(result)
if not vals:
logging.error(u"No data.")
return
txt_chains = list()
txt_nodes = list()
for key_c in vals:
txt_chains.append(key_c)
for key_n in vals[key_c].keys():
txt_nodes.append(key_n)
if vals[key_c][key_n][u"vals"]:
vals[key_c][key_n][u"nr"] = len(vals[key_c][key_n][u"vals"])
vals[key_c][key_n][u"mean"] = \
round(mean(vals[key_c][key_n][u"vals"]) / 1000000, 1)
vals[key_c][key_n][u"stdev"] = \
round(stdev(vals[key_c][key_n][u"vals"]) / 1000000, 1)
txt_nodes = list(set(txt_nodes))
def sort_by_int(value):
"""Makes possible to sort a list of strings which represent integers.
:param value: Integer as a string.
:type value: str
:returns: Integer representation of input parameter 'value'.
:rtype: int
"""
return int(value)
txt_chains = sorted(txt_chains, key=sort_by_int)
txt_nodes = sorted(txt_nodes, key=sort_by_int)
chains = [i + 1 for i in range(len(txt_chains))]
nodes = [i + 1 for i in range(len(txt_nodes))]
data = [list() for _ in range(len(chains))]
for chain in chains:
for node in nodes:
try:
val = vals[txt_chains[chain - 1]][txt_nodes[node - 1]][u"mean"]
except (KeyError, IndexError):
val = None
data[chain - 1].append(val)
# Color scales:
my_green = [[0.0, u"rgb(235, 249, 242)"], [1.0, u"rgb(45, 134, 89)"]]
my_blue = [[0.0, u"rgb(236, 242, 248)"], [1.0, u"rgb(57, 115, 172)"]]
my_grey = [[0.0, u"rgb(230, 230, 230)"], [1.0, u"rgb(102, 102, 102)"]]
hovertext = list()
annotations = list()
text = (u"Test: {name}<br>"
u"Runs: {nr}<br>"
u"Thput: {val}<br>"
u"StDev: {stdev}")
for chain, _ in enumerate(txt_chains):
hover_line = list()
for node, _ in enumerate(txt_nodes):
if data[chain][node] is not None:
annotations.append(
dict(x=node + 1,
y=chain + 1,
xref=u"x",
yref=u"y",
xanchor=u"center",
yanchor=u"middle",
text=str(data[chain][node]),
font=dict(size=14, ),
align=u"center",
showarrow=False))
hover_line.append(
text.format(
name=vals[txt_chains[chain]][txt_nodes[node]][u"name"],
nr=vals[txt_chains[chain]][txt_nodes[node]][u"nr"],
val=data[chain][node],
stdev=vals[txt_chains[chain]][
txt_nodes[node]][u"stdev"]))
hovertext.append(hover_line)
traces = [
plgo.Heatmap(x=nodes,
y=chains,
z=data,
colorbar=dict(
title=plot.get(u"z-axis", u""),
titleside=u"right",
titlefont=dict(size=16),
tickfont=dict(size=16, ),
tickformat=u".1f",
yanchor=u"bottom",
y=-0.02,
len=0.925,
),
showscale=True,
colorscale=my_green,
text=hovertext,
hoverinfo=u"text")
]
for idx, item in enumerate(txt_nodes):
# X-axis, numbers:
annotations.append(
dict(x=idx + 1,
y=0.05,
xref=u"x",
yref=u"y",
xanchor=u"center",
yanchor=u"top",
text=item,
font=dict(size=16, ),
align=u"center",
showarrow=False))
for idx, item in enumerate(txt_chains):
# Y-axis, numbers:
annotations.append(
dict(x=0.35,
y=idx + 1,
xref=u"x",
yref=u"y",
xanchor=u"right",
yanchor=u"middle",
text=item,
font=dict(size=16, ),
align=u"center",
showarrow=False))
# X-axis, title:
annotations.append(
dict(x=0.55,
y=-0.15,
xref=u"paper",
yref=u"y",
xanchor=u"center",
yanchor=u"bottom",
text=plot.get(u"x-axis", u""),
font=dict(size=16, ),
align=u"center",
showarrow=False))
# Y-axis, title:
annotations.append(
dict(x=-0.1,
y=0.5,
xref=u"x",
yref=u"paper",
xanchor=u"center",
yanchor=u"middle",
text=plot.get(u"y-axis", u""),
font=dict(size=16, ),
align=u"center",
textangle=270,
showarrow=False))
updatemenus = list([
dict(x=1.0,
y=0.0,
xanchor=u"right",
yanchor=u"bottom",
direction=u"up",
buttons=list([
dict(args=[{
u"colorscale": [
my_green,
],
u"reversescale": False
}],
label=u"Green",
method=u"update"),
dict(args=[{
u"colorscale": [
my_blue,
],
u"reversescale": False
}],
label=u"Blue",
method=u"update"),
dict(args=[{
u"colorscale": [
my_grey,
],
u"reversescale": False
}],
label=u"Grey",
method=u"update")
]))
])
try:
layout = deepcopy(plot[u"layout"])
except KeyError as err:
logging.error(f"Finished with error: No layout defined\n{repr(err)}")
return
layout[u"annotations"] = annotations
layout[u'updatemenus'] = updatemenus
try:
# Create plot
plpl = plgo.Figure(data=traces, layout=layout)
# Export Plot
logging.info(f" Writing file {plot[u'output-file']}.html")
ploff.plot(plpl,
show_link=False,
auto_open=False,
filename=f"{plot[u'output-file']}.html")
except PlotlyError as err:
logging.error(f" Finished with error: {repr(err)}".replace(
u"\n", u" "))
return
def plotly_heatmap(self):
"""plot_heatmap, but with plotly output"""
fig = go.Figure()
cols = self.plots_per_row if self.numplots > self.plots_per_row else self.numplots
rows = np.ceil(self.numplots / float(cols)).astype(int)
fig['layout'].update(title=self.plot_title)
domainWidth = .9 / cols
domainHeight = .9 / rows
bufferHeight = 0.0
if rows > 1:
bufferHeight = 0.1 / (rows - 1)
else:
domainHeight = 1.0
bufferWidth = 0.0
if cols > 1:
bufferWidth = 0.1 / (cols - 1)
else:
domainWidth = 1.0
data = []
annos = []
zmin = np.inf
zmax = -np.inf
for i in range(self.numplots):
row = rows - i / self.plots_per_row - 1
col = i % self.plots_per_row
if self.per_group:
title = self.hm.matrix.group_labels[i]
else:
title = self.hm.matrix.sample_labels[i]
base = row * (domainHeight + bufferHeight)
domain = [base, base + domainHeight]
titleY = base + domainHeight
xanchor = 'x{}'.format(i + 1)
yanchor = 'y{}'.format(i + 1)
visible = False
if col == 0:
visible = True
fig['layout']['yaxis{}'.format(i + 1)] = {'domain': domain, 'anchor': xanchor, 'visible': visible}
base = col * (domainWidth + bufferWidth)
domain = [base, base + domainWidth]
titleX = base + 0.5 * domainWidth
fig['layout']['xaxis{}'.format(i + 1)] = {'domain': domain, 'anchor': yanchor}
annos.append({'yanchor': 'bottom', 'xref': 'paper', 'xanchor': 'center', 'yref': 'paper', 'text': title, 'y': titleY, 'x': titleX, 'font': {'size': 16}, 'showarrow': False})
mat = []
labels = []
for j in range(self.numlines):
if self.per_group:
row, col = i, j
else:
row, col = j, i
sub_matrix = self.hm.matrix.get_matrix(row, col)
if self.per_group:
label = sub_matrix['sample']
else:
label = sub_matrix['group']
labels.append(label)
mat.append(np.ma.__getattribute__(self.averagetype)(sub_matrix['matrix'], axis=0))
if np.min(mat[-1]) < zmin:
zmin = np.min(mat[-1])
if np.max(mat[-1]) > zmax:
zmax = np.max(mat[-1])
totalWidth = len(mat[-1])
trace = go.Heatmap(name=title, z=mat, x=range(totalWidth + 1), y=labels, xaxis=xanchor, yaxis=yanchor)
data.append(trace)
# Add ticks
xticks, xtickslabel = self.getTicks(i)
if np.ceil(max(xticks)) != float(totalWidth):
tickscale = float(totalWidth) / max(xticks)
xticks_use = [x * tickscale for x in xticks]
else:
xticks_use = xticks
xticks_use = [np.ceil(x) for x in xticks_use]
fig['layout']['xaxis{}'.format(i + 1)].update(tickmode='array', tickvals=xticks_use, ticktext=xtickslabel, tickangle=self.label_rotation)
# Adjust color scale limits
for i, trace in enumerate(data):
zminUse = zmin
zmaxUse = zmax
if self.y_min[i % len(self.y_min)] is not None:
zminUse = self.y_min[i % len(self.y_min)]
if self.y_max[i % len(self.y_max)] is not None:
zmaxUse = self.y_max[i % len(self.y_max)]
trace.update(zmin=zminUse, zmax=zmaxUse)
fig['data'] = data
fig['layout']['annotations'] = annos
py.plot(fig, filename=self.out_file_name, auto_open=False)
m2.append(males2[0][i])
m3.append(males3[0][i])
f1.append(females1[0][i])
f2.append(females2[0][i])
f3.append(females3[0][i])
for j in range(0, 11):
cor[i][j] = cor1[i][j] + cor2[i][j] + cor3[i][j]
candidates = [
'NDA', 'MLP', 'EM', 'BH', 'NA', 'PP', 'JC', 'JL', 'JLM', 'FA', 'FF'
]
trace0 = go.Heatmap(z=cor,
x=candidates,
y=candidates,
colorbar=dict(x=0.47, y=0.81, len=0.4),
colorscale='Viridis',
name='All VTs')
trace1 = go.Heatmap(z=cor1,
x=candidates,
y=candidates,
colorbar=dict(y=0.81, len=0.4),
name='VT1')
trace2 = go.Heatmap(z=cor2,
x=candidates,
y=candidates,
colorbar=dict(x=0.47, y=0.19, len=0.4),
name='VT2')
trace3 = go.Heatmap(z=cor3,
x=candidates,
y=candidates,
plty.iplot(fig, filename='bar-direct-labels')
# with kfold
base_predictions_train = pd.DataFrame({
'RandomeForest': rf_oof_train.ravel(),
'ExtraTrees': et_oof_train.ravel(),
'AdaBoost': ada_oof_train.ravel(),
'GradientBoost': gb_oof_train.ravel(),
})
base_predictions_train.head()
data = [
go.Heatmap(
z=base_predictions_train.astype(float).corr().values,
x=base_predictions_train.columns.values,
y=base_predictions_train.columns.values,
colorscale='Portland',
showscale=True,
# reverscale=True
)
]
plty.iplot(data, filename='labelled-heatmap')
x_train = np.concatenate((et_oof_train, rf_oof_train, ada_oof_train, gb_oof_train, svc_oof_train), axis=1)
x_test = np.concatenate((et_oof_test, rf_oof_test, ada_oof_test, gb_oof_test, svc_oof_test), axis=1)
gbm = xgb.XGBClassifier(
learning_rate=0.02,
n_estimators=2000,
max_depth=4,
min_child_weight=2,
H.set_dict_options(options)
H
#il est nécessaire d'installer 'plotly' si ce n'est pas déjà fait
# Connexion à l'API avec un compte existant
import plotly
plotly.tools.set_credentials_file(username='******',
api_key='P7zRkAFgDxfcFdcGwzTP')
# Chargement des librairies
import plotly.plotly as py
import plotly.graph_objs as go
# Création de la 'heatmap'
heatmap = go.Heatmap(z=acteurs_large.as_matrix(),
x=acteurs_large.columns,
y=acteurs_large.index,
colorscale='Jet')
data = [heatmap]
py.iplot(data, filename='ATMO-heatmap')
# Définition des accès
user = '******'
password = '******'
machine = '172.20.152.200'
mySQLengine = create_engine("mysql://%s:%s@%s/?charset=utf8" %
(user, password, machine))
# Ici, on execute directement du SQL: "USE XXX"
mySQLengine.execute("USE `BDD_Alexandre`;")
# Ecriture des données dataframe -> mySQL
# Après avoir supprimé (si existante) la table 'people'
# df = pd.DataFrame(data=evaluated_response_data,
# columns=['apple', 'google', 'amazon', 'facebook', 'microsoft'])
# df.head()
doc = collection.find_one()
heat_map_df = pd.DataFrame(data=[doc for doc in collection.find()])
heat_map_df.drop(columns=['_id'], inplace=True)
heat_map_df.set_index('METRIC NAME', inplace=True)
heat_map_df.head()
response_df = pd.Dataframe()
app = dash.Dash()
app.layout = html.Div([
dcc.Graph(id='heap_map',
figure={
'data': [
go.Heatmap(z=heat_map_df.values.tolist(),
x=heat_map_df.columns.tolist(),
y=heat_map_df.index.tolist())
],
'layout':
go.Layout(title='Heap Map of Company Response')
})
])
if __name__ == '__main__':
PORT = 8000
ADDRESS = '127.0.0.1'
app.run_server(port=PORT, host=ADDRESS)
import plotly.offline as pyo
import plotly.graph_objs as go
import pandas as pd
df = pd.read_csv('dados/2010YumaAZ.csv')
data = [go.Heatmap(x=df['DAY'],y=df['LST_TIME'], z=df['T_HR_AVG'].values.tolist())]
layout = go.Layout(title='Yuma')
fig = go.Figure(data=data, layout=layout)
pyo.plot(fig, filename='HTML/heatmaps_2.html')
#API = raw_input("Enter your plotly API key: ")
#py.sign_in(username, API)
with Dataset("tmin.nc", "r") as tmin:
tempMinJ = tmin.variables["tmin"][0]
tempMinA = tmin.variables["tmin"][91]
tempMinY = tmin.variables["tmin"][182]
tempMinO = tmin.variables["tmin"][274]
#min temps at certain days of year
#of course I would loop this through every day of the year if I wasn't just doing four for now...
#plot the heatmaps for tmin with the same zmin and zmax values
data1 = [go.Heatmap(
z=tempMinJ,
zmin=-7,
zmax=27,
)]
data2 = [go.Heatmap(
z=tempMinA,
zmin=-7,
zmax=27,
)]
data3 = [go.Heatmap(
z=tempMinY,
zmin=-7,
zmax=27,
)]
def plot_pfb(filename,
decimation=None,
low_pass=None,
backend='matplotlib',
output_filename=None,
start_time=None,
end_time=None,
auto_open=True,
**kwargs):
'''
Plot the output of a PFB acquisition as an heatmap.
:retrurn Name of the file.
'''
final_filename = ""
filename = format_filename(filename)
parameters = global_parameter()
parameters.retrive_prop_from_file(filename)
ant = parameters.get_active_rx_param()
try:
usrp_number = kwargs['usrp_number']
except KeyError:
usrp_number = None
if len(ant) > 1:
print_error("multiple RX devices not yet supported")
return
if parameters.get(ant[0], 'wave_type')[0] != "NOISE":
print_warning(
"The file selected does not have the PFB acquisition tag. Errors may occour"
)
fft_tones = parameters.get(ant[0], 'fft_tones')
rate = parameters.get(ant[0], 'rate')
channel_width = rate / fft_tones
decimation = parameters.get(ant[0], 'decim')
integ_time = fft_tones * max(decimation, 1) / rate
rf = parameters.get(ant[0], 'rf')
if start_time is not None:
start_time *= effective_rate
else:
start_time = 0
if end_time is not None:
end_time *= effective_rate
try:
front_end = kwargs['front_end']
except KeyError:
front_end = None
samples, errors = openH5file(filename,
ch_list=None,
start_sample=start_time,
last_sample=end_time,
usrp_number=usrp_number,
front_end=front_end,
verbose=False,
error_coord=True)
if output_filename is None:
output_filename = "PFB_waterfall_" + get_timestamp()
y_label = np.arange(len(samples[0]) /
fft_tones) / (rate / (fft_tones * max(1, decimation)))
x_label = (rf + (np.arange(fft_tones) - fft_tones / 2) *
(rate / fft_tones)) / 1e6
title = "PFB acquisition form file %s" % filename
subtitle = "Channel width %.2f kHz; Frame integration time: %.2e s" % (
channel_width / 1.e3, integ_time)
with warnings.catch_warnings():
# it's very likely to do some division by 0 in the log10
warnings.simplefilter("ignore")
z = 20 * np.log10(np.abs(samples[0]))
try:
z_shaped = np.roll(np.reshape(z, (len(z) / fft_tones, fft_tones)),
-fft_tones / 2,
axis=1)
except ValueError as msg:
print_warning("Error while plotting pfb spectra: " + str(msg))
cut = len(z) - len(z) / fft_tones * fft_tones
z = z[:-cut]
print_debug("Cutting last data (%d samples) to fit" % cut)
# z_shaped = np.roll(np.reshape(z,(len(z)/fft_tones,fft_tones)),fft_tones/2,axis = 1)
z_shaped = np.roll(np.reshape(z, (len(z) / fft_tones, fft_tones)),
-fft_tones / 2,
axis=1)
# pl.plot(z_shaped.T, alpha = 0.1, color = "k")
# pl.show()
if backend == 'matplotlib':
fig, ax = pl.subplots(
nrows=2,
ncols=1,
) # sharex=True)
try:
fig.set_size_inches(kwargs['size'][0], kwargs['size'][1])
except KeyError:
fig.set_size_inches(10, 16)
ax[0].set_xlabel("Channel [MHz]")
ax[0].set_ylabel("Time [s]")
ax[0].set_title(title + "\n" + subtitle)
ax[0].set_xlim((min(x_label), max(x_label)))
imag = ax[0].imshow(
z_shaped,
aspect='auto',
interpolation='nearest',
extent=[min(x_label),
max(x_label),
min(y_label),
max(y_label)])
# fig.colorbar(imag)#,ax=ax[0]
for zz in z_shaped[::100]:
ax[1].plot(x_label, zz, color='k', alpha=0.1)
ax[1].set_xlabel("Channel [MHz]")
ax[1].set_ylabel("Power [dBm]")
ax[1].set_xlim((min(x_label), max(x_label)))
# ax[1].set_title("Trace stack")
final_filename = output_filename + '.png'
fig.savefig(final_filename)
pl.close()
if backend == 'plotly':
data = [
go.Heatmap(
z=z_shaped,
x=x_label,
y=y_label,
colorscale='Viridis',
)
]
layout = go.Layout(title=title + "<br>" + subtitle,
xaxis=dict(title="Channel [MHz]"),
yaxis=dict(title="Time [s]"))
fig = go.Figure(data=data, layout=layout)
final_filename = output_filename + ".html"
plotly.offline.plot(fig)
plotly.offline.plot(fig, filename=final_filename, auto_open=auto_open)
return final_filename
query_command = {
'metadata.timestamp': {
'$gt': start_time,
'$lt': end_time
}
}
runs = collection.find(query_command)
res_stats = list()
timestamps = list()
for document in runs:
res_stats.append(int(document['metadata']['success']))
timestamps.append(document['metadata']['timestamp'])
z.append(res_stats)
y.append(collection_name)
res_heatmap = go.Heatmap(z=z,
x=timestamps,
y=y,
zmin=0,
zmax=1,
reversescale=True,
showscale=False)
_res_stats_graph = dcc.Graph(figure=go.Figure(data=[res_heatmap],
layout={'title': 'Run Results'}),
id='res_heatmap')
pass_fail_stats = html.Div([_res_stats_graph])
def plotly_scatter(self,
plot_filename,
corr_matrix,
plot_title='',
minXVal=None,
maxXVal=None,
minYVal=None,
maxYVal=None):
"""Make the scatter plot of a matrix with plotly"""
n = self.matrix.shape[1]
self.matrix = self.matrix
fig = go.Figure()
domainWidth = 1. / n
annos = []
for i in range(n):
x = domainWidth * (i + 1)
y = 1 - (domainWidth * i + 0.5 * domainWidth)
anno = dict(text=self.labels[i],
showarrow=False,
xref='paper',
yref='paper',
x=x,
y=y,
xanchor='right',
yanchor='middle')
annos.append(anno)
data = []
zMin = np.inf
zMax = -np.inf
for x in range(n):
xanchor = 'x{}'.format(x + 1)
base = x * domainWidth
domain = [base, base + domainWidth]
if x > 0:
base = 1 - base
fig['layout']['xaxis{}'.format(x + 1)] = dict(
domain=domain,
range=[minXVal, maxXVal],
anchor='free',
position=base)
for y in range(0, n):
yanchor = 'y{}'.format(y + 1)
if x == 1:
base = 1 - y * domainWidth
domain = [base - domainWidth, base]
fig['layout']['yaxis{}'.format(y + 1)] = dict(
domain=domain,
range=[minYVal, maxYVal],
side='right',
anchor='free',
position=1.0)
if x > y:
vector1 = self.matrix[:, x]
vector2 = self.matrix[:, y]
Z, xEdges, yEdges = np.histogram2d(vector1,
vector2,
bins=50)
Z = np.log10(Z)
if np.min(Z) < zMin:
zMin = np.min(Z)
if np.max(Z) > zMax:
zMax = np.max(Z)
name = '{}={:.2f}'.format(self.corr_method, corr_matrix[x,
y])
trace = go.Heatmap(z=Z,
x=xEdges,
y=yEdges,
showlegend=False,
xaxis=xanchor,
yaxis=yanchor,
name=name,
showscale=False)
data.append(trace)
# Fix the colorbar bounds
for trace in data:
trace.update(zmin=zMin, zmax=zMax)
data[-1]['colorbar'].update(title="log10(instances per bin)",
titleside="right")
data[-1].update(showscale=True)
fig['data'] = data
fig['layout'].update(title=plot_title,
showlegend=False,
annotations=annos)
offline.plot(fig, filename=plot_filename, auto_open=False)
def plot_slot_avail(node_info, sec_since_epoch=None):
"""Plot "slot" availability.
Parameters
----------
node_info : pandas.DataFrame
sec_since_epoch : numeric, optional
If provided, a date-time stamp will be displayed on the plots
indicating the time the information displayed was generated.
"""
if DEBUG:
t0 = time.time()
print(">> plot_slot_avail")
updated_at = ""
if sec_since_epoch is not None:
datetime_stamp, _ = make_datetime_stamp(sec_since_epoch,
human_readable=True)
updated_at = "Updated {}".format(datetime_stamp)
slot_cores = np.array(
[1, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 24, 28, 32, 40])
slot_mem = 2**np.arange(0, 11)
cluster_node_info = OrderedDict([
("ACI CPU", node_info.query('cluster == "aci"')),
(
"CyberLAMP CPU-Only",
node_info.query('(cluster == "cyberlamp") & (subgroup != "phi")'),
),
(
"CyberLAMP Single-GPU",
node_info.query('(cluster == "cyberlamp") & (subgroup != "phi")'),
),
])
for cluster_name, cni in cluster_node_info.items():
print(">> Working on {} cluster".format(cluster_name))
# TODO: cyberlamp has different CPUs available on the 1-GPU nodes for
# GPU vs. non-GPU jobs. Knowing which cores are "taken" (or if it's
# just a count, not particular cores that are dedicatd to job type) is
# not possible as the pbsnodes parsing script stands (might need data
# from another source about actual jobs that are running on each node
# and what they call out).
sc = slot_cores[slot_cores <= cni.cores.max()]
sm = slot_mem[slot_mem <= cni.mem_tot.max()]
if "gpu" in cluster_name.lower():
sg = 1
sc = sc[sc <= 20]
sm = sm[sm <= 242]
else:
sg = 0
slots = compute_open_slots(node_info=cni,
slot_cores=sc,
slot_mem=sm,
slot_gpus=sg)
if np.all(slots <= 3):
log_slots = slots
else:
with np.errstate(divide="ignore"):
log_slots = np.log10(slots)
neginf_mask = np.isinf(log_slots)
neginf_fill_val = -np.max(log_slots) / 6
log_slots[neginf_mask] = neginf_fill_val
min_slots = slots.min()
max_slots = slots.max()
lin_tickvals = np.array(
[1, 3, 10, 30, 100, 300, 1e3, 3e3, 1e4, 3e4, 1e5, 3e5, 1e6],
dtype=int)
lin_tickvals = lin_tickvals[(lin_tickvals > min_slots)
& (lin_tickvals < max_slots)]
if np.all(slots <= 3):
lin_tickvals = np.array([0, max_slots], dtype=int)
tickvals = lin_tickvals
else:
if len(lin_tickvals) < 3:
order = np.floor(np.log10(min_slots))
lin_tickvals = np.arange(
10**np.floor(np.log10(min_slots)),
max_slots + 1,
10**order,
dtype=int,
)
lin_tickvals = np.array(sorted(
set(lin_tickvals.tolist() + [max_slots])),
dtype=int)
tickvals = np.log10(lin_tickvals)
lin_tickvals = sorted(set([0] + lin_tickvals.tolist()))
tickvals = [neginf_fill_val] + tickvals.tolist()
ticktext = ["{:d}".format(tv) for tv in lin_tickvals]
text = np.empty_like(slots, dtype=str).tolist()
for (core_i, cores), (mem_i, mem) in product(enumerate(sc),
enumerate(sm)):
if cores == 1:
ct = "core"
else:
ct = "cores"
val = slots[core_i, mem_i]
if val == 1:
slt = "slot"
else:
slt = "slots"
text[core_i][
mem_i] = "{:d} {:s} & {:d} GiB mem : {:d} {:s}".format(
cores, ct, mem, val, slt)
# print(f"slots\n{slots}")
# print(f"log_slots\n{log_slots}")
# print(f"tickvals\n{tickvals}")
trace = graph_objs.Heatmap(
z=log_slots,
x=[str(m) + " GiB" for m in sm],
y=[str(sc[0]) + " core"] + [str(c) + " cores" for c in sc[1:]],
zsmooth=False,
xgap=1,
ygap=1,
colorscale="Viridis",
colorbar=dict(
outlinewidth=0,
tickvals=tickvals,
ticktext=ticktext,
),
hoverinfo="text",
text=text,
)
data = [trace]
layout = graph_objs.Layout(
title="{} Job Slots Available".format(cluster_name),
annotations=graph_objs.Annotations([
graph_objs.Annotation(
x=0.5,
y=1.07,
showarrow=False,
text=updated_at,
xref="paper",
yref="paper",
),
]),
xaxis=dict(ticks="", nticks=36),
yaxis=dict(ticks=""),
)
fig = graph_objs.Figure(data=data, layout=layout)
plotly.plot(fig, filename=cluster_name, auto_open=False)
if DEBUG:
print(">> plot_slot_avail:", time.time() - t0)
def plotlyMatrix(hm,
outFilename,
yMin=[None], yMax=[None],
zMin=[None], zMax=[None],
showSummaryPlot=False,
cmap=None, colorList=None, colorBarPosition='side',
perGroup=False,
averageType='median', yAxisLabel='', xAxisLabel='',
plotTitle='',
showColorbar=False,
label_rotation=0.0):
label_rotation *= -1.0
if colorBarPosition != 'side':
sys.error.write("Warning: It is not currently possible to have multiple colorbars with plotly!\n")
nRows = hm.matrix.get_num_groups()
nCols = hm.matrix.get_num_samples()
if perGroup:
nRows, nCols = nCols, nRows
profileHeight = 0.0
profileBottomBuffer = 0.0
if showSummaryPlot:
profileHeight = 0.2
profileBottomBuffer = 0.05
profileSideBuffer = 0.
profileWidth = 1. / nCols
if nCols > 1:
profileSideBuffer = 0.1 / (nCols - 1)
profileWidth = 0.9 / nCols
dataSummary = []
annos = []
fig = go.Figure()
fig['layout'].update(title=plotTitle)
xAxisN = 1
yAxisN = 1
# Summary plots at the top (if appropriate)
if showSummaryPlot:
yMinLocal = np.inf
yMaxLocal = -np.inf
for i in range(nCols):
xanchor = 'x{}'.format(xAxisN)
yanchor = 'y{}'.format(yAxisN)
xBase = i * (profileSideBuffer + profileWidth)
yBase = 1 - profileHeight
xDomain = [xBase, xBase + profileWidth]
yDomain = [yBase, 1.0]
for j in range(nRows):
if perGroup:
mat = hm.matrix.get_matrix(i, j)
xTicks, xTicksLabels = hm.getTicks(i)
label = mat['sample']
else:
mat = hm.matrix.get_matrix(j, i)
xTicks, xTicksLabels = hm.getTicks(j)
label = mat['group']
if j == 0:
fig['layout']['xaxis{}'.format(xAxisN)] = dict(domain=xDomain, anchor=yanchor, range=[0, mat['matrix'].shape[1]], tickmode='array', tickvals=xTicks, ticktext=xTicksLabels, tickangle=label_rotation)
fig['layout']['yaxis{}'.format(yAxisN)] = dict(anchor=xanchor, domain=yDomain)
trace = plotly_single(mat['matrix'], averageType, colorList[j], label)[0]
trace.update(xaxis=xanchor, yaxis=yanchor, legendgroup=label)
if min(trace['y']) < yMinLocal:
yMinLocal = min(trace['y'])
if max(trace['y']) > yMaxLocal:
yMaxLocal = max(trace['y'])
if i == 0:
trace.update(showlegend=True)
dataSummary.append(trace)
# Add the column label
if perGroup:
title = hm.matrix.group_labels[i]
else:
title = hm.matrix.sample_labels[i]
titleX = xBase + 0.5 * profileWidth
annos.append({'yanchor': 'bottom', 'xref': 'paper', 'xanchor': 'center', 'yref': 'paper', 'text': title, 'y': 1.0, 'x': titleX, 'font': {'size': 16}, 'showarrow': False})
xAxisN += 1
yAxisN += 1
# Adjust y-bounds as appropriate:
for i in range(1, yAxisN):
yMinUse = yMinLocal
if yMin[(i - 1) % len(yMin)] is not None:
yMinUse = yMin[(i - 1) % len(yMin)]
yMaxUse = yMaxLocal
if yMax[(i - 1) % len(yMax)] is not None:
yMaxUse = yMax[(i - 1) % len(yMax)]
fig['layout']['yaxis{}'.format(i)].update(range=[yMinUse, yMaxUse])
fig['layout']['yaxis1'].update(title=yAxisLabel)
# Add the heatmap
dataHeatmap = []
zMinLocal = np.inf
zMaxLocal = -np.inf
heatmapWidth = 1. / nCols
heatmapSideBuffer = 0.0
if nCols > 1:
heatmapWidth = .9 / nCols
heatmapSideBuffer = 0.1 / (nCols - 1)
heatmapHeight = 1.0 - profileHeight - profileBottomBuffer
for i in range(nCols):
xanchor = 'x{}'.format(xAxisN)
xBase = i * (heatmapSideBuffer + heatmapWidth)
# Determine the height of each heatmap, they have no buffer
lengths = [0.0]
for j in range(nRows):
if perGroup:
mat = hm.matrix.get_matrix(i, j)
else:
mat = hm.matrix.get_matrix(j, i)
lengths.append(mat['matrix'].shape[0])
fractionalHeights = heatmapHeight * np.cumsum(lengths).astype(float) / np.sum(lengths).astype(float)
xDomain = [xBase, xBase + heatmapWidth]
fig['layout']['xaxis{}'.format(xAxisN)] = dict(domain=xDomain, anchor='free', position=0.0, range=[0, mat['matrix'].shape[1]], tickmode='array', tickvals=xTicks, ticktext=xTicksLabels, title=xAxisLabel)
# Start adding the heatmaps
for j in range(nRows):
if perGroup:
mat = hm.matrix.get_matrix(i, j)
label = mat['sample']
start = hm.matrix.group_boundaries[i]
end = hm.matrix.group_boundaries[i + 1]
else:
mat = hm.matrix.get_matrix(j, i)
label = mat['group']
start = hm.matrix.group_boundaries[j]
end = hm.matrix.group_boundaries[j + 1]
regs = hm.matrix.regions[start:end]
regs = [x[2] for x in regs]
yanchor = 'y{}'.format(yAxisN)
yDomain = [heatmapHeight - fractionalHeights[j + 1], heatmapHeight - fractionalHeights[j]]
visible = False
if i == 0:
visible = True
fig['layout']['yaxis{}'.format(yAxisN)] = dict(domain=yDomain, anchor=xanchor, visible=visible, title=label, tickmode='array', tickvals=[], ticktext=[])
if np.min(mat['matrix']) < zMinLocal:
zMinLocal = np.min(mat['matrix'])
if np.max(mat['matrix']) < zMaxLocal:
zMaxLocal = np.max(mat['matrix'])
trace = go.Heatmap(z=np.flipud(mat['matrix']),
y=regs[::-1],
xaxis=xanchor,
yaxis=yanchor,
showlegend=False,
name=label,
showscale=False)
dataHeatmap.append(trace)
yAxisN += 1
xAxisN += 1
if showColorbar:
dataHeatmap[-1].update(showscale=True)
dataHeatmap[-1]['colorbar'].update(len=heatmapHeight, y=0, yanchor='bottom', ypad=0.0)
# Adjust z bounds and colorscale
for trace in dataHeatmap:
zMinUse = zMinLocal
zMaxUse = zMaxLocal
if zMin[0] is not None:
zMinUse = zMin[0]
if zMax[0] is not None:
zMaxUse = zMax[0]
trace.update(zmin=zMinUse, zmax=zMaxUse, colorscale=convertCmap(cmap[0], vmin=zMinUse, vmax=zMaxUse))
dataSummary.extend(dataHeatmap)
fig['data'] = dataSummary
fig['layout']['annotations'] = annos
py.plot(fig, filename=outFilename, auto_open=False)
'Unrecovered': 'sum'
}).reset_index()
data_bubblechart = [
go.Scatter(x=bubble_df['Recovered'],
y=bubble_df['Unrecovered'],
text=bubble_df['Country'],
mode='markers',
marker=dict(size=bubble_df['Confirmed'] / 200,
color=bubble_df['Confirmed'] / 200,
showscale=True))
]
# Heatmap
data_heatmap = [
go.Heatmap(x=df2['Day'],
y=df2['WeekofMonth'],
z=df2['Recovered'].values.tolist(),
colorscale='Jet')
]
# Layout
app.layout = html.Div(children=[
html.H1(children='Python Dash',
style={
'textAlign': 'center',
'color': '#ef3e18'
}),
html.Div('Web dashboard for Data Visualization using Python',
style={'textAlign': 'center'}),
html.Div('Coronavirus COVID-19 Global Cases - 1/22/2020 to 3/17/2020',
style={'textAlign': 'center'}),
html.Br(),
#Scatter Plot
import plotly.express as px
fig = px.scatter(drug_info_dataframe.query("year==2018"), x="state", y="count",
size="year", color="city",
hover_name="year",size_max=20,title='State Wise Drug Death in USA 2018')
fig.show()
# Heat Map
import numpy as np
import plotly.graph_objects as go
fig = go.Figure(data=go.Heatmap(
z=np.array(drug_info_dataframe_2['avg_drug_score']),
x=np.array(drug_info_dataframe_2['age_bin']),
y=np.array(drug_info_dataframe_2['state']),
hoverongaps = False))
fig.show()
# Interactive Visulaization
import numpy as np
import pandas as pd
import plotly.graph_objects as go
from ipywidgets import widgets
year = widgets.IntSlider(
value=1,
min=2012,
def heatmap(df,
colname,
plot_title='default',
resolution='day',
aggfunc='mean',
zmin='default',
zmax='default',
zsmooth=None,
reversescale=False,
colorscale='RdBu',
colorbar_title='Units',
height=600,
width=1000,
plot=True,
asFigure=False,
layoutupdate=False):
'''take df, x, y, z (optional) columns as integers and
return a plotly heatmap'''
if type(df) == pd.core.series.Series:
df = pd.DataFrame(df)
x, plt_title = colname_to_int(df, colname)
dfpivot = pd.DataFrame(df.loc[:, plt_title])
dfpivot['hour'] = dfpivot.index.hour
dfpivot['day'] = dfpivot.index.map(lambda x: x.strftime('%b-%d'))
dfpivot['week'] = dfpivot.index.week
dfpivot['month'] = dfpivot.index.month
dfpivot['dayofyear'] = dfpivot.index.dayofyear
dfpivot.columns = [''.join(x) for x in dfpivot.columns]
dfpivot = dfpivot.pivot_table(
values=dfpivot.columns[0],
index='hour',
columns=['dayofyear', 'week', 'day', 'month'],
aggfunc=aggfunc)
trace = go.Heatmap(x=[x[2] for x in dfpivot.columns],
y=dfpivot.index.values,
z=dfpivot.values,
zsmooth=zsmooth,
reversescale=reversescale,
colorscale=colorscale,
colorbar=dict(title=colorbar_title))
if zmin != 'default':
if zmax == 'default':
raise ValueError(
"Error: If entering zmin or zmax, BOTH must be specified.")
trace.update(zmin=zmin)
trace.update(zmax=zmax)
if plot_title == 'default':
title = str(plt_title) + " Annual Heatmap"
else:
title = plot_title
layout = dict(title=title, height=height, width=width)
data = [trace]
fig = dict(data=data, layout=layout)
fig['layout'].update({
'font': {
'family': 'Futura LT BT, monospace',
'size': 14
},
'title': title,
'titlefont': {
'size': 24
},
'xaxis': {
'title': 'Day of Year'
},
'yaxis': {
'title': 'Hour of Day'
}
})
if layoutupdate:
layout.update(layoutupdate)
if lv(plotly.__version__) >= lv('3.5'):
fig['data'][0]['colorbar']['title'].update({'side': 'right'})
else:
fig['data'][0]['colorbar'].update({'titleside': 'right'})
if plot:
py.iplot(fig, config=config)
if asFigure:
return fig
}
y = []
x = []
sPi = '0' + str(pi)
sPi = sPi.replace('.', '')
# create plot
StartTime = datetime.now()
print(StartTime, '\t\tBegin Plotting PI()')
for index in range(0, len(sPi)): # Build the plot series
# print(sPi[index])
if index == 0:
x.append(0)
y.append(0)
else:
x.append(float(xDict[int(sPi[index])]) + x[index - 1])
y.append(float(yDict[int(sPi[index])]) + y[index - 1])
EndTime = datetime.now()
print(EndTime, '\t\tPI() Plot Completed')
print(EndTime - StartTime, '\t\tTotal Plot Time\r\n')
answer = input('Plot? ')
answer = answer.lower()
if answer.startswith('y'):
# Plot the image
trace1 = go.Heatmap([(x), (y)])
data = [trace]
py.iplot(data)
#######
# Heatmap of temperatures for Yuma, Arizona
######
import plotly.offline as pyo
import plotly.graph_objs as go
import pandas as pd
df = pd.read_csv('../data/2010YumaAZ.csv')
data = [
go.Heatmap(x=df['DAY'],
y=df['LST_TIME'],
z=df['T_HR_AVG'].values.tolist(),
colorscale='Jet')
]
layout = go.Layout(title='Hourly Temperatures, June 1-7, 2010 in<br>\
Yuma, AZ USA')
fig = go.Figure(data=data, layout=layout)
pyo.plot(fig, filename='Yuma.html')
