def do_execute_direct(self, code):
if self._first:
self._first = False
self.handle_plot_settings()
self._do_execute_direct(self._setup)
self._skipping = os.name != 'nt'
super(ScilabKernel, self).do_execute_direct(code, self.Print)
if self.plot_settings.get('backend', None) == 'inline':
plot_dir = tempfile.mkdtemp()
self._make_figs(plot_dir)
width, height = 0, 0
for fname in os.listdir(plot_dir):
filename = os.path.join(plot_dir, fname)
try:
if fname.lower().endswith('.svg'):
im = SVG(filename)
if ',' in self.plot_settings['size']:
size = self.plot_settings['size']
width, height = size.split(',')
width, height = int(width), int(height)
im.data = self._fix_svg_size(im.data,
size=(width, height))
else:
im = Image(filename)
self.Display(im)
except Exception as e:
self.Error(e)
def _handle_svg(self, filename):
"""
Handle special considerations for SVG images.
"""
# Gnuplot can create invalid characters in SVG files.
with codecs.open(filename, 'r', encoding='utf-8',
errors='replace') as fid:
data = fid.read()
im = SVG(data=data)
try:
im.data = self._fix_svg_size(im.data)
except Exception:
pass
return im
def do_execute_direct(self, code):
if self._first:
self._first = False
if sys.platform == "darwin":
self.plot_settings["format"] = "svg"
self.handle_plot_settings()
super(OctaveKernel, self).do_execute_direct(self._setup)
if os.name != "nt":
msg = (
"may not be able to display plots properly "
"without gnuplot, please install it "
"(gnuplot-x11 on Linux)"
)
try:
subprocess.check_call(["gnuplot", "--version"])
except subprocess.CalledProcessError:
self.Error(msg)
resp = super(OctaveKernel, self).do_execute_direct(code)
if self.plot_settings.get("backend", None) == "inline":
plot_dir = tempfile.mkdtemp()
self._make_figs(plot_dir)
width, height = 0, 0
for fname in os.listdir(plot_dir):
filename = os.path.join(plot_dir, fname)
try:
if fname.lower().endswith(".svg"):
im = SVG(filename)
if "," in self.plot_settings["size"]:
size = self.plot_settings["size"]
width, height = size.split(",")
width, height = int(width), int(height)
im.data = self._fix_gnuplot_svg_size(im.data, size=(width, height))
else:
im = Image(filename)
self.Display(im)
except Exception as e:
self.Error(e)
return resp or None
def do_execute_direct(self, code):
if self._first:
self._first = False
self.handle_plot_settings()
super(OctaveKernel, self).do_execute_direct(self._setup)
if os.name != 'nt':
msg = ('may not be able to display plots properly '
'without gnuplot, please install it '
'(gnuplot-x11 on Linux)')
try:
subprocess.check_call(['gnuplot', '--version'])
except subprocess.CalledProcessError:
self.Error(msg)
super(OctaveKernel, self).do_execute_direct(code, self.Print)
if self.plot_settings.get('backend', None) == 'inline':
plot_dir = tempfile.mkdtemp()
self._make_figs(plot_dir)
width, height = 0, 0
for fname in os.listdir(plot_dir):
filename = os.path.join(plot_dir, fname)
try:
if fname.lower().endswith('.svg'):
im = SVG(filename)
size = self.plot_settings['size']
if not (isinstance(size, tuple)):
size = 560, 420
width, height = size
width, height = int(width), int(height)
im.data = self._fix_gnuplot_svg_size(im.data,
size=(width, height))
else:
im = Image(filename)
self.Display(im)
except Exception as e:
import traceback
traceback.print_exc(file=sys.__stderr__)
self.Error(e)
counts["best_move"].append(best_move)
counts["best_move_score"].append(next_scores[0])
best_move_score_diff = abs(next_scores[0] - score)
counts["best_move_score_diff"].append(best_move_score_diff)
counts["best_move_score_diff_category"].append(
categorize_best_move_score_diff(best_move_score_diff, best_move, actual_move))
else:
counts["best_move"].append("-")
counts["best_move_score"].append(0)
counts["best_move_score_diff"].append(0)
# push actual move to the board again
board.push(move)
exportBoardSVG('output/' + filename + '/images/', len(counts["san"]))
SVG(chess.svg.board(board=board, size=400))
# use plotly to visualize the chess game in a line graph
import plotly
# import plotly.plotly as py
import plotly.graph_objs as go
import plotly.io as pio
# plotly.tools.set_credentials_file(username='******', api_key='••••••••••')
plotly.offline.init_notebook_mode(connected=True)
checkcolor = ['red' if i else 'white' for i in counts["is_check"]]
checknr = [i for (i, s) in enumerate(counts["is_check"]) if s]
bubble = [s / 2 for s in counts["move_count"]]
# best = [np.log(s+1) for s in counts["bestdiff"]]
# best = [s for s in counts["bestdiff"]]
def draw(self, filename=None, opts={}, **kwargs):
"""
filename specifies the name of the file to produce. If None,
the schematic is displayed on the screen.
Note, if using Jupyter, then need to first issue command %matplotlib inline
kwargs include:
label_ids: True to show component ids
label_values: True to display component values
draw_nodes: True to show all nodes, False to show no nodes,
'primary' to show primary nodes,
'connections' to show nodes that connect more than two components,
'all' to show all nodes
label_nodes: True to label all nodes, False to label no nodes,
'primary' to label primary nodes,
'alpha' to label nodes starting with a letter,
'pins' to label nodes that are pins on a chip,
'all' to label all nodes
style: 'american', 'british', or 'european'
scale: schematic scale factor, default 1.0
node_spacing: spacing between component nodes, default 2.0
cpt_size: size of a component, default 1.5
oversample: oversampling factor for png or pdf files
help_lines: distance between lines in grid, default 0.0 (disabled)
debug: True to display debug information
"""
for key, val in opts.items():
if key not in kwargs or kwargs[key] is None:
kwargs[key] = val
def in_ipynb():
try:
ip = get_ipython()
cfg = ip.config
kernapp = cfg['IPKernelApp']
# Check if processing ipynb file.
if 'connection_file' in kernapp:
return True
elif kernapp and kernapp[
'parent_appname'] == 'ipython-notebook':
return True
else:
return False
except (NameError, KeyError):
return False
if not self.hints:
raise RuntimeWarning('No schematic drawing hints provided!')
png = 'png' in kwargs and kwargs.pop('png')
svg = 'svg' in kwargs and kwargs.pop('svg')
if not png and not svg:
png = True
if in_ipynb() and filename is None:
if png:
from IPython.display import Image, display_png
pngfilename = tmpfilename('.png')
self.tikz_draw(pngfilename, **kwargs)
# Create and display PNG image object.
# There are two problems:
# 1. The image metadata (width, height) is ignored
# when the ipynb file is loaded.
# 2. The image metadata (width, height) is not stored
# when the ipynb file is written non-interactively.
display_png(
Image(filename=pngfilename,
width=self.width * 100,
height=self.height * 100))
return
if svg:
from IPython.display import SVG, display_svg
svgfilename = tmpfilename('.svg')
self.tikz_draw(svgfilename, **kwargs)
# Create and display SVG image object.
# Note, there is a problem displaying multiple SVG
# files since the later ones inherit the namespace of
# the first ones.
display_svg(
SVG(filename=pngfilename,
width=self.width * 100,
height=self.height * 100))
return
if filename is None:
filename = tmpfilename('.png')
self.tikz_draw(filename=filename, **kwargs)
display_matplotlib(filename)
return
self.tikz_draw(filename=filename, **kwargs)
def plot_model(model):
return SVG(model_to_dot(model).create(prog='dot', format='svg'))
def viewms(objs,wo3d=F,nmr=8,molSize=(100,80),wlgd=T,wlgd_text='smiles', labels=None, \
wia=T,woH=T,kekulize=T,atsh=[],bsh=[],colors=[],filename=None):
"""
vars
atsh : atoms to be highlighted (list)
"""
mols, strings = draw_rdkit.str2mol(objs, woH=woH)
mcs = []
_labels = []
nat = 0
nm = len(mols)
smis = []
for mol in mols:
labels_i = [
chemical_symbols[ai.GetAtomicNum()] + '%d' % (ai.GetIdx() + 1)
for ai in mol.GetAtoms()
]
nat += len(labels_i)
_labels += labels_i
mc = Chem.Mol(mol.ToBinary())
if kekulize:
try:
Chem.Kekulize(mc)
except:
mc = Chem.Mol(mol.ToBinary())
c2d = F
if mc.GetNumConformers():
if wo3d:
c2d = T
else:
c2d = T
if c2d: rdDepictor.Compute2DCoords(mc)
smis.append(Chem.MolToSmiles(mc))
mcs.append(mc)
#print('labels=',labels)
molsPerRow = nmr #5
subImgSize = molSize
nRows = len(mols) // molsPerRow
if len(mols) % molsPerRow:
nRows += 1
fsize = (molsPerRow * subImgSize[0], nRows * subImgSize[1])
dw, dh = 0, 0
drawer = rdMolDraw2D.MolDraw2DSVG(fsize[0] + dw, fsize[1] + dh,
subImgSize[0], subImgSize[1])
opts = drawer.drawOptions()
## [bad news] opts seems to be able to accept only the first `na1 labels (where na1=len(mols[0]))
#for i in range(nat):
# opts.atomLabels[i] = labels[i]
if labels:
labs = labels
else:
if wlgd_text in ['id']:
labs = [str(i + 1) for i in range(nm)] #rows*cols) ]
elif wlgd_text in [
'smiles',
]:
labs = smis
elif wlgd_text in ['file', 'filename']:
assert len(fs) > 1, '#ERROR: input objs are not files!!'
labs = [fi.split('/')[-1] for fi in fs]
else:
raise Exception('invalid `wlgd_text')
drawer.DrawMolecules(mcs, legends=labs)
drawer.FinishDrawing()
svg = drawer.GetDrawingText()
# It seems that the svg renderer used doesn't quite hit the spec.
# Here are some fixes to make it work in the notebook, although I think
# the underlying issue needs to be resolved at the generation step
if filename is not None:
open(filename + '.svg', 'w').write(svg)
try:
import cairosvg
cairosvg.svg2pdf(url='%s.svg' % filename,
write_to='%s.pdf' % filename)
except:
raise Exception(
'#ERROR: svg to pdf failed, as cairosvg seems to be not installed?'
)
return SVG(svg.replace('svg:', ''))
output_dim=embedding_dim,
input_length=400))
# mlp.add(layers.Flatten())
mlp.add(layers.GlobalMaxPool1D())
mlp.add(layers.Dense(128, activation='relu', kernel_regularizer = regularizers.l2(l = 0.001)))
mlp.add(layers.Dropout(0.5))
mlp.add(layers.Dense(64, activation='relu', kernel_regularizer = regularizers.l2(l = 0.001)))
mlp.add(layers.Dropout(0.5))
mlp.add(layers.Dense(7, activation='softmax'))
mlp.compile(optimizer='adam',
loss='categorical_crossentropy',
metrics=['accuracy'],
)
mlp.summary()
SVG(model_to_dot(mlp, show_shapes=True, show_layer_names=False).create(prog='dot', format='svg'))
"""#### Measure performance of MLP"""
import matplotlib.pyplot as plt
plt.style.use('ggplot')
def plot_history(history):
acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']
x = range(1, len(acc) + 1)
plt.figure(figsize=(12, 5))
plt.subplot(1, 2, 1)
acc = accuracy_score(testY, predY)
#confusion matrix calculation
cfm = confusion_matrix(testY, predY, labels=[0, 1])
print('Confusion Matrix:\n', cfm)
print('\nAccuracy: ', acc)
# calculating per class precision and per class recall
precision = []
recall = []
precision.append(cfm[0, 0] / (cfm[0, 0] + cfm[1, 0]))
precision.append(cfm[1, 1] / (cfm[0, 1] + cfm[1, 1]))
recall.append(cfm[0, 0] / (cfm[0, 0] + cfm[0, 1]))
recall.append(cfm[1, 1] / (cfm[1, 0] + cfm[1, 1]))
print('\nPrecision: ', precision)
print('\nRecall: ', recall)
# Decision tree visualization
# Note: Decision tree is huge,
# so some times it feels like there is no output, to see the whole tree please scroll down
print("\nTree Visualization")
graph = Source(
tree.export_graphviz(
clf,
out_file=None,
feature_names=['Occupation', 'Marital_Status', 'Age', 'Gender']))
SVG(graph.pipe(format='svg'))
tech_rets = closing_df.pct_change()
tech_rets.head()
# In[59]:
sns.jointplot('GOOG', 'GOOG', tech_rets, kind="scatter", color='seagreen')
# In[60]:
sns.jointplot('GOOG', 'MSFT', tech_rets, kind="scatter")
# In[61]:
from IPython.display import SVG
SVG(url=
'http://upload.wikimedia.org/wikipedia/commons/d/d4/Correlation_examples2.svg'
)
# In[63]:
#correlation between different stocks
sns.pairplot(tech_rets.dropna())
# In[66]:
return_fig = sns.PairGrid(tech_rets.dropna())
return_fig.map_upper(plt.scatter, color='purple')
return_fig.map_lower(sns.kdeplot, cmap='cool_d')
return_fig.map_diag(plt.hist, bins=30)
# In[67]:
movehistory =[]
board = chess.Board()
boardvalue = init_evaluate_board()
i = 1
x = 5
if move == 0:
print (board)
if color == "black":
while not board.is_game_over(claim_draw=True):
ponder = 0
mov = selectmove(x)
make_move(mov)
print (mov)
SVG(chess.svg.board(board=board,size=400,lastmove=mov))
print (board)
playermove = input ("what is your move?")
board.push_san(playermove)
print(board)
if color == "white":
while not board.is_game_over(claim_draw=True):
ponder = 1
playermove = input ("what is your move?")
board.push_san(playermove)
print (board)
ponder = 0
mov = selectmove(x)
make_move(mov)
print (mov)
def draw(self):
return SVG(self.graph.draw(format='svg', prog='dot'))
from IPython.display import SVG from keras.utils.vis_utils import model_to_dot from keras.utils import plot_model import pydot import os os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' #启动神经网络 network = models.Sequential() #添加使用ReLU激活函数的全连接层 network.add(layers.Dense(units=16, activation="relu", input_shape=(10, ))) #添加使用ReLU激活函数的全连接层 network.add(layers.Dense(units=16, activation="relu")) #添加使用sigmoid激活函数的全连接层 network.add(layers.Dense(units=1, activation="sigmoid")) #可视化网络结构 #SVG(model_to_dot(network,show_shapes=False).create(prog="dot",format="svg")) #keras提供了工具函数用于快速可视化神经网络 #model_to_dot可以显示一个神经网络 # show_shapes 参数指定是否展示输入输出的形状，可以帮助我们调试网络 #将可视化后的网络结构图保存为文件 #plot_model(network,show_shapes=True,to_file="network.png") #展示一个更加简单的网络模型 SVG(model_to_dot(network, show_shapes=False).create(prog="dot", format="svg")) plot_model(network, show_shapes=True, to_file="network.png")
def plot_network(network):
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
SVG(model_to_dot(network).create(prog='dot', format='svg'))
def show_model(model):
return SVG(
model_to_dot(model, show_shapes=True,
show_layer_names=False).create(prog='dot', format='svg'))
def svg(self):
from IPython.display import SVG
return SVG(self._repr_svg_())
def viz_model_architecture(model):
"""Visualize model architecture in Jupyter notebook."""
display(SVG(model_to_dot(model).create(prog='dot', format='svg')))
at https://www.debuggingbook.org/html/Tours.html
'''
# Allow to use 'from . import <module>' when run as script (cf. PEP 366)
if __name__ == '__main__' and __package__ is None:
__package__ = 'debuggingbook'
# Tours through the Book
# ======================
if __name__ == '__main__':
print('# Tours through the Book')
if __name__ == '__main__':
from IPython.display import SVG
if __name__ == '__main__':
SVG(filename='PICS/Sitemap.svg')
## The Pragmatic Programmer Tour
## -----------------------------
if __name__ == '__main__':
print('\n## The Pragmatic Programmer Tour')
## Lessons Learned
## ---------------
if __name__ == '__main__':
print('\n## Lessons Learned')
etree.Element("rect", {
"width": "100%",
"height": "100%",
"fill": "white"
}),
panel_1a,
panel_1b,
panel_1c,
panel_1dleft,
panel_1dright,
panel_1a_label,
panel_1b_label,
panel_1c_label,
panel_1d_label,
])
display(SVG(figure_1.to_str()))
# save generated SVG files
figure_1.save("output/figure_1.svg")
# ## Figure 2
# Create panels for figure 2
panel_2a = make_figure_panel("fig2A.svg",
scale_x_input=2,
scale_y_input=2,
x_loc=30,
y_loc=10)
panel_2b = make_figure_panel(
"../human_cancer_analysis/gene_ranking_logFC.svg",
scale_x_input=0.6,
# coding: utf-8 # Conv2Dを使ったCNNの例 # CNN Model 3 - zero padding # Kerasとその他ライブラリをインポート from keras.models import Sequential, Model from keras.layers import Conv2D from keras.utils import np_utils # SVGの表示に必要なライブラリのインポート from IPython.display import SVG from keras.utils.vis_utils import model_to_dot # Windows の場合は以下を追加 import os os.environ["PATH"] += os.pathsep + 'C:/Program Files (x86)/Graphviz2.38/bin/' # パディングつき畳み込みニューラルネットワークのモデルを作成 model = Sequential() model.add(Conv2D(filters=3, kernel_size=(3, 3), input_shape=(6, 6, 1), padding='same', name='Conv2D_1')) # SVG形式でモデルを表示 SVG(model_to_dot(model, show_shapes=True).create(prog='dot', format='svg'))
def plotm(model):
display(
SVG(
model_to_dot(model, show_shapes=True).create(prog='dot',
format='svg')))
def svg(self, line, cell):
"""Render the cell as an SVG literal"""
display(SVG(cell))
def render_model(cobra_model,
background_template=None,
custom_css=None,
figure_id=None,
hide_unused=None,
hide_unused_cofactors=None,
inactive_alpha=1.,
figsize=None,
label=None,
fontsize=None,
default_flux_width=2.5,
flux_dict=None):
""" Render a cobra.Model object in the current window
Parameters:
background_template:
filename for an SVG to render behind the flux figure. Useful for
compartments or layout guides.
custom_css:
Additional CSS to embed in the figure. Use HTML inspector to show
labels and classes applied to reactions and metabolites.
figure_id:
Each figure in the page requires a unique ID, which can be passed or
generated automatically.
hide_unused:
whether or not to show metabolites and reactions with zero flux.
hide_unused_cofactors:
similar to hide_unused, but only hide cofactor nodes for reactions with
0 flux.
inactive_alpha:
Alpha value with which to color reactions and nodes without any carried
flux. Defaults to 1.
figsize:
size, in pixels, of the generated SVG window. Defaults to 1024x768.
fontsize:
text size, in pt. Defaults to 12
default_flux_width:
If reaction fluxes are missing, the default thickness to use for
connecting arrows.
flux_dict:
A dictionary-like object containing the desired fluxes for each
reaction in the model
"""
# Increment figure counter
# Get figure name and JSON string for the cobra model
if not figure_id:
render_model._fignum += 1
figure_id = 'd3flux{:0>3d}'.format(render_model._fignum)
if not figsize:
figsize = (1028, 768)
modeljson = create_model_json(cobra_model, flux_dict)
if not hide_unused:
hide_unused = "false"
else:
hide_unused = "true"
if not hide_unused_cofactors:
hide_unused_cofactors = "false"
else:
hide_unused_cofactors = "true"
# Handle custom CSS
if not custom_css:
custom_css = ''
if not fontsize:
fontsize = 12
# Handle background template
if not background_template:
background_svg = ''
no_background = "true"
else:
from IPython.display import SVG
background_svg = SVG(background_template).data
no_background = "false"
# Initialize the jinja templates
env = Environment(loader=FileSystemLoader(
os.path.join(os.path.dirname(d3flux.__file__), 'templates')))
template_css = env.get_template('network_style.css')
template_html = env.get_template('output_template.html')
template_js = env.get_template('d3flux.js')
# Render the jinja templates with the given variables
css = template_css.render(inactive_alpha=inactive_alpha)
js = template_js.render(figure_id=figure_id,
modeljson=modeljson,
no_background=no_background,
hide_unused=hide_unused,
hide_unused_cofactors=hide_unused_cofactors,
figwidth=figsize[0],
figheight=figsize[1],
fontsize=fontsize,
css=compress(css + custom_css),
default_flux_width=default_flux_width)
html = template_html.render(figure_id=figure_id,
background_svg=background_svg,
javascript_source=js)
# compile and return HTML
return HTML(html)
import numpy as np
np.random.seed(123) # for reproducibility
import os; os.environ['KERAS_BACKEND'] = 'theano'
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Convolution2D, MaxPooling2D
from keras.utils import np_utils
from keras.datasets import mnist
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
from keras.models import load_model
os.environ["PATH"] += os.pathsep + 'C:/anaconda3/Lib/site-packages/graphviz-2.38/bin'
(X_train, y_train), (X_test, y_test) = mnist.load_data()
model = load_model('models/mnistCNN.h5')
SVG(model_to_dot(model).create(prog='dot', format='svg'))
import plotly.graph_objects as go
fig = go.Figure()
fig.add_trace(go.Scatter(
x=data['a'],
y=data['b']
))
fig.show()
# %% [markdown]
# # Show image
# %%
import plotly.io as pio
from IPython.display import SVG, display, Image
img_bytes = pio.to_image(fig, format="svg")
display(SVG(img_bytes))
# %%
fig.write_image('out.png')
# %% [markdown]
# # Export json for Plotly viewer
# %%
pio.write_json(fig,'demo_plotly.json')
# %% [markdown]
# # SQL access
# %%
# python database api specification v2.0
def viewms(objs, nmr=8, molSize=[100,80], woH=T, wo3d=F, wlgd_text='smiles',\
wlgd=T, wia=F, filename=None, tdir='/tmp/'):
"""
vars
====================
nmr: num of mol per row
woH: without Hydrogens displayed, T/F
wo3d: without 3d coordinates? T/F
wlgd: with legend displayed? T/F
wlgd_text: 'smiles' or 'idx'
wia: with idx of atom shown? T/F
"""
mols, smis, fs = draw_oechem.str2mol(objs, woH=woH, wo3d=wo3d)
nm = len(mols)
cols = nmr #4 if nm < nmr else nmr # num mol per row
rows = int(nm / cols) if nm % cols == 0 else int(nm / cols) + 1
image = oedepict.OEImage(cols * molSize[0], rows * molSize[1])
grid = oedepict.OEImageGrid(image, rows, cols)
opts = oedepict.OE2DMolDisplayOptions(grid.GetCellWidth(), \
grid.GetCellHeight(), oedepict.OEScale_AutoScale)
if wlgd_text in ['id']:
labs = [str(i + 1) for i in range(nm)] #rows*cols) ]
elif wlgd_text in [
'smiles',
]:
labs = smis
elif wlgd_text in ['file', 'filename']:
assert len(fs) > 1, '#ERROR: input objs are not files!!'
labs = [fi.split('/')[-1] for fi in fs]
else:
raise Exception('invalid `wlgd_text')
for i in range(nm):
mol = mols[i]
#print('mol=',mol)
oedepict.OEPrepareDepiction(mol)
irow, icol = int(i / cols) + 1, i % cols + 1
cell = grid.GetCell(irow, icol)
if wlgd:
mol.SetTitle(labs[i])
opts.SetAtomVisibilityFunctor(
oechem.OEIsTrueAtom()) # explicitly set the default
font = oedepict.OEFont()
font.SetSize(16)
font.SetStyle(oedepict.OEFontStyle_Normal) #_Bold)
font.SetColor(oechem.OEDarkGreen)
opts.SetAtomPropLabelFont(font)
if wia:
opts.SetAtomPropertyFunctor(oedepict.OEDisplayAtomIdx())
opts.SetAtomLabelFont(font)
#opts.SetAtomLabelFontScale(1.5)
opts.SetAtomPropLabelFont(font) # atom idx
opts.SetAtomPropLabelFontScale(2) # atom idx
opts.SetTitleLocation(oedepict.OETitleLocation_Bottom)
opts.SetTitleHeight(10.0)
disp = oedepict.OE2DMolDisplay(mol, opts)
clearbackground = True
oedepict.OERenderMolecule(cell, disp, not clearbackground)
save = F # save as pdf
if filename is not None:
tf = filename + '.svg'
save = T
else:
tf = tpf.NamedTemporaryFile(dir=tdir).name + '.svg'
oedepict.OEWriteImage(tf, image)
if save:
try:
import cairosvg
cairosvg.svg2pdf(url='%s.svg' % filename,
write_to='%s.pdf' % filename)
except:
raise Exception(
'#ERROR: svg to pdf failed, as cairosvg seems to be not installed?'
)
return SVG(tf)
def image(self):
"""Returns the result as an SVG image for output in jupyter"""
from IPython.display import SVG
return SVG(self.result["svgUrl"])
'artifacts/model.pb': '1Ap3eWHWwAyw_3wOmy237AJF3pWQRnG3_', # pb format
'artifacts/model.nnet': '1HzfGxhKrw9PpeA1cMsexC4FcWv5OPdtB', # nnet format
'artifacts/scaler.pkl': '10EkqHQ3aqEYAxbLS4Q4LRWJ1byNCvAcf', # scaler object
'artifacts/onehot.pkl': '1SeED9m_TeyqtmHRgDe_kd9HVmn2K1hh8' # onehot object
}
# upload all of the artifacts to drive
# for fname,driveid in artifacts.items():
# overwrite_gdrive_file(driveid, fname)
"""
## Visualization
"""
!mkdir -p images
display(tf.keras.utils.plot_model(model, to_file='images/model.png', show_shapes=True, show_layer_names=True, expand_nested=True, dpi=1200, rankdir='LR'))
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
svg_plot = tf.keras.utils.model_to_dot(model, show_shapes=False, show_layer_names=True, rankdir='LR')
SVG(svg_plot.create(prog='dot', format='svg'))
# with open('images/model.svg', 'w') as f:
# f.write(svg_plot.decode('utf-8'))
svg_plot.write_svg('images/model.svg')
# Commented out IPython magic to ensure Python compatibility.
# %tensorboard --logdir $tensorboard_logs --host localhost --port 6006
def make_svg(self):
from IPython.display import SVG
return SVG(self.make_dot().create_svg())
h = Dense(hidden_size, activation='relu')(x)
r = Dense(output_size, activation='sigmoid')(h)
# 构建模型，给定模型优化参数
autoencoder = Model(inputs=x, outputs=r)
autoencoder.compile(optimizer='adam', loss='mse')
# --------------------- 3、构建单层自编码器模型 ---------------------
# --------------------- 4、模型可视化 ---------------------
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
# 可视化模型
SVG(model_to_dot(autoencoder).create(prog='dot', format='svg'))
# --------------------- 4、模型可视化 ---------------------
# --------------------- 5、训练 ---------------------
# 设定peochs和batch_size大小
epochs = 5
batch_size = 128
# 训练模型
history = autoencoder.fit(X_train,
X_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
import chess
import chess.svg
from IPython.display import SVG
board = chess.Board()
board
board.push_san("e4")
board
SVG(chess.svg.board(board=board))
from keras.models import model_from_json
from keras.utils.vis_utils import plot_model, model_to_dot
from IPython.display import SVG, display_svg
# load model
json_file = open('model_100.json', 'r')
loaded_model_json = json_file.read()
json_file.close()
loaded_model = model_from_json(loaded_model_json)
# load model weights
loaded_model.load_weights("model_100.h5")
plot_model(loaded_model,
to_file='model_plot.png',
show_shapes=True,
show_layer_names=True)
display(SVG(model_to_dot(loaded_model).create(prog='dot', format='svg')))
eeg_df.head()
eeg_df=eeg_df.iloc[:,1:]
filted_df=generate_samples(eeg_df,489).reset_index()
y=filted_df['group']
X=filted_df.drop(columns=['index','group'])
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=1)
X_valid, X_train = X_train[:100], X_train[100:]
y_valid, y_train = y_train[:100], y_train[100:]
X_train,X_valid,X_test=transformat_define(X_train),transformat_define(X_valid),transformat_define(X_test)
y_train,y_valid,y_test=transformat_define(y_train),transformat_define(y_valid),transformat_define(y_test)
X_train=X_train.astype('float32')
X_test=X_test.astype('float32')
onehot_train = keras.utils.to_categorical(y_train, num_classes=2)
onehot_test=keras.utils.to_categorical(y_test,num_classes=2)
model.add(Dense(30,activation='relu',input_shape=(72,)))
model.add(Dense(20,activation='relu'))
model.add(Dense(10,activation='relu'))
model.add(Dense(2,activation='softmax'))
model.summary()
from keras.utils.vis_utils import plot_model
plot_model(model,to_file="keras_dnn.png")
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
SVG(model_to_dot(model).create(prog="dot",format='svg'))
from keras import optimizers
sgd=optimizers.SGD(lr=0.01,momentum=0.9)
model.compile(optimizer=sgd,loss="categorical_crossentropy",metrics=['accuracy'])
model.fit(x=X_train,y=onehot_train,epochs=50,batch_size=20)
model.evaluate(X_test,onehot_test,batch_size=128)
