def plot_label_heatmap(samples, bars_y = 30):
angles = np.linspace(0., np.pi, len(samples[0]['y']))
min_x = np.amin(angles)
max_x = np.amax(angles)
radii = np.zeros((len(samples), len(angles)))
for i, s in enumerate(samples):
radii[i] = s['y']
min_y = np.amin(radii)
max_y = np.amax(radii)
heatmap_array = np.zeros((bars_y, len(angles)))
for i in xrange(np.shape(radii)[0]):
for j in xrange(np.shape(radii)[1]):
dest_j = j
dest_i = round((radii[i,j] - min_y) / (max_y - min_y) * bars_y)-1
heatmap_array[dest_i, dest_j] += 1
indices = np.linspace(min_y, max_y, bars_y)
indices = ["%.1f" % i for i in indices]
columns = np.linspace(0., 180., len(angles))
columns = ["%.1f" % i for i in columns]
heatmap_array = np.flipud(heatmap_array)
heatmap_frame = pandas.DataFrame(data=heatmap_array, index=reversed(indices), columns=columns)
f = sns.heatmap(heatmap_frame)
plt.subplots_adjust(top=0.9)
plt.title("Labels Heatmap")
sns.axlabel("Angle", "Radius")
sns.plt.show(f)
def plot_kmeans_components(self, fig1, gs, kmeans_clusters, clrs, plot_title='Hb', num_subplots=1,
flag_separate=1, gridspecs=[0, 0]):
with sns.axes_style('darkgrid'):
if flag_separate:
ax1 = fig1.add_subplot(2, 1, num_subplots)
else:
ax1 = eval('fig1.add_subplot(gs' + gridspecs + ')')
plt.gca().set_color_cycle(clrs)
for ii in xrange(0, size(kmeans_clusters, 1)):
plt.plot(kmeans_clusters[:, ii], lw=4, label=str(ii))
plt.locator_params(axis='y', nbins=4)
sns.axlabel("Time (seconds)", "a.u")
ax1.legend(prop={'size': 14}, loc='center left', bbox_to_anchor=(1, 0.5), ncol=1, fancybox=True,
shadow=True)
plt.title(plot_title, fontsize=14)
plt.ylim((min(kmeans_clusters) - 0.0001,
max(kmeans_clusters) + 0.0001))
plt.xlim((0, size(kmeans_clusters, 0)))
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
self.plot_vertical_lines_onset()
self.plot_vertical_lines_offset()
def plot_joint(self, cmap="BuGn"):
"""Plot the current joint distribution P(p, I)."""
pal = sns.color_palette(cmap, 256)
lc = pal[int(.7 * 256)]
bg = pal[0]
fig = plt.figure(figsize=(7, 7))
gs = plt.GridSpec(6, 6)
p_lim = self.p_grid.min(), self.p_grid.max()
I_lim = self.I_grid.min(), self.I_grid.max()
ax1 = fig.add_subplot(gs[1:, :-1])
ax1.set(xlim=p_lim, ylim=I_lim)
ax1.contourf(self.p_grid, self.I_grid, self.pI.T, 30, cmap=cmap)
sns.axlabel("$p$", "$I$", size=16)
ax2 = fig.add_subplot(gs[1:, -1], axis_bgcolor=bg)
ax2.set(ylim=I_lim)
ax2.plot(self.pI.sum(axis=0), self.I_grid, c=lc, lw=3)
ax2.set_xticks([])
ax2.set_yticks([])
ax3 = fig.add_subplot(gs[0, :-1], axis_bgcolor=bg)
ax3.set(xlim=p_lim)
ax3.plot(self.p_grid, self.pI.sum(axis=1), c=lc, lw=3)
ax3.set_xticks([])
ax3.set_yticks([])
def plot_scores(matched_signals, unique_clrs, ind, stimulus_on_time, stimulus_off_time):
######Plot mean signals according to color and boxplot of number of pixels in each plane
sns.tsplot(np.array(matched_signals[ind].clr_grped_signal), linewidth=3, ci=95, err_style="ci_band", color=unique_clrs[ind])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
plot_vertical_lines_onset(stimulus_on_time)
plot_vertical_lines_offset(stimulus_off_time)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
def plot_prediction_error(name, clf, X, y):
plt.figure()
cv = KFold(X.shape[0], 5, shuffle=True)
predicted = cross_val_predict(clf, X, y, cv=cv)
print("%.3f = mean squared error" % mean_squared_error(y, predicted))
sns.regplot(x=y[:1000], y=predicted[:1000])
sns.axlabel("actual", "predicted")
plt.savefig("plot_validation_" + name + ".png")
def tshist(ax,X,Y):
B=25
assert(X.size==Y.shape[1])
img=np.zeros((B,X.size))
for t in range(X.size):
hc,be=np.histogram(Y[:,t],bins=B,range=(-1,1))
img[::-1,t]=hc.astype(np.float)/Y.shape[0]
sns.heatmap(img,ax=ax,cmap='gray')
sns.axlabel('X','A')
def grafico_l2(conjunto, xl=None, yl=None, titulox="", tituloy="", titulo="", filename="", tamanho=5):
a = np.array(conjunto[0].map(_dic_cruzes))
b = np.array(conjunto[1].map(_dic_cruzes))
c = DataFrame([a, b]).transpose()
c.columns = ["A", "B"]
sns.lmplot("A", "B", c, x_jitter=0.2, y_jitter=0.3, size=tamanho)
plt.title(titulo, fontsize=16)
sns.axlabel(titulox, tituloy, fontsize=fontetamanho)
plt.savefig(filename)
def grafico_j(conjunto, xl=None, yl=None, titulox="", tituloy="", filename="", tamanho=5, titulo=""):
a = np.array(conjunto[0].map(float))
b = np.array(conjunto[1].map(float))
c = DataFrame([a, b]).transpose()
c.columns = ["A", "B"]
g = sns.jointplot(
"A", "B", c, xlim=xl, ylim=yl, kind="reg", stat_func=stats.spearmanr, marginal_kws={"bins": 25}, size=tamanho
)
# marginal_kws={"bins": 297}
sns.axlabel(titulox, tituloy, fontsize=fontetamanho)
plt.title(titulo, y=1.22, fontsize=20)
plt.savefig(filename, bbox_inches="tight", format="png")
def plot_normed_intensities(normed_intensities, path=""):
"""
Plot the normalized intensities to show that the distributions
overlap post-normalization
"""
for col in normed_intensities.columns:
sns.kdeplot(np.log2(normed_intensities[col]).dropna(), label=col, shade=True)
plt.legend(bbox_to_anchor=(1.3, 0.5), loc="center right")
sns.despine()
sns.axlabel("Log$_2$ Normalized Intensities", "Density")
plt.tight_layout()
plt.savefig(os.path.join(path, "normed_intensities_wcl.png"), dpi=300)
def plot_scores(self, fig1, gs, matched_signals, unique_clrs, plot_title='Habenula', gridspecs='[0,0]'):
with sns.axes_style('dark'):
ax1 = eval('fig1.add_subplot(gs' + gridspecs + ')')
for ind in range(0, size(unique_clrs, 0)):
sns.tsplot(array(matched_signals[ind].clr_grped_signal), linewidth=5, ci=95, err_style="ci_band",
color=unique_clrs[ind])
ax1.locator_params(axis='y', nbins=4)
sns.axlabel("Time (seconds)", "a.u")
plt.title(plot_title, fontsize=14)
self.plot_vertical_lines_onset()
self.plot_vertical_lines_offset()
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
def plot_ica_components(ica_components_plot, colors_ica, ax1,stimulus_on_time, stimulus_off_time):
########### Plot components ##################
for ii in xrange(0, np.size(ica_components_plot, 1)):
plt.plot(ica_components_plot[:,ii], color=colors_ica[ii])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
A = []
for ii in xrange(0,np.size(ica_components_plot, 1)):
A = np.append(A, [str(ii+1)])
ax1.legend(A, loc=4)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
plot_vertical_lines_onset(stimulus_on_time)
plot_vertical_lines_offset(stimulus_off_time)
def plot_matchedpixels(fig1, gs, matched_pixels, unique_clrs, gridspecs=[0, 0]):
ax1 = eval('fig1.add_subplot(gs' + gridspecs + ')')
with sns.axes_style("darkgrid"):
for ii in xrange(0, size(matched_pixels, 0)):
plt.plot(ii + 1, transpose(matched_pixels[ii, :]), 'o', color=unique_clrs[ii], markersize=10)
plt.xlim([0, size(matched_pixels, 0) + 1])
for ii in range(0, size(unique_clrs, 0)):
plt.plot(repeat(ii + 1, size(matched_pixels, 1)), transpose(matched_pixels[ii, :]), 's',
color=unique_clrs[ii], markersize=10, markeredgecolor='k', markeredgewidth=2)
x = range(0, size(unique_clrs, 0)+1)
labels = [str(e) for e in x]
plt.xticks(x, labels, rotation='vertical')
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True)
def graph_func2(perrault):
sns.set( font_scale=.8)
sns.axlabel('Color', 'Frequency' )
#colors from http://www.color-hex.com and wikipedia
flatui = ["#4b0082","#ffd700", "#e6e6fa", "#ffff00", "#FF2400", "#ff6eb4", "#d2b48c", "#ff00ff", "#0000ff", "#C8A2C8",
"#800080", "#FF007F", "#FD3F92", "#000000", "#dc143c", "#CCCCFF", "#ffffff", "#ff0000", "#631919", "#fffff0",
"#ffa500","#730000", "#808000","#00ffff","#c0c0c0","#808080", "#7fffd4", "#808080", "#008000", "#f5f5dc",
"#329999", "#f0ffff", "#FFEF00"]
custom_palette = sns.color_palette(flatui)
colors = perrault[0]
occurences = perrault[1]
ax2 = sns.barplot(colors, occurences, palette = custom_palette)
fig2 = ax2.get_figure()
for item in ax2.get_xticklabels():
item.set_rotation(45)
sns.plt.title('Color Word Frequencies in Charles Perrault Stories')
fig2.savefig('perrault_chart.png')
fig2.clf()
def plot_pca_components(pca_components,ax1,stimulus_on_time, stimulus_off_time,required_pcs):
########### Plot components ##################
for ii in range(np.size(pca_components,1)):
if ii in required_pcs:
plt.plot(pca_components[:,ii],'--', linewidth=4)
else:
plt.plot(pca_components[:,ii])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
A = []
for ii in xrange(0,np.size(pca_components, 1)):
A = np.append(A, [str(ii)])
ax1.legend(A, loc=4)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
plot_vertical_lines_onset(stimulus_on_time)
plot_vertical_lines_offset(stimulus_off_time)
def plot_boxplot(fig2, matched_pixels, unique_clrs):
#### Plot Boxplot of number of pixels
## Dont plot boxplot if there is only one Z
if np.size(matched_pixels,1) == 1:
with sns.axes_style("darkgrid"):
for ii in xrange(0,np.size(matched_pixels,0)):
fig2 = plt.plot(ii+1,np.transpose(matched_pixels[ii,:]),'o', color=unique_clrs[ii])
plt.xlim([0,np.size(matched_pixels,0)+1])
else:
fig2 = sns.boxplot(np.transpose(matched_pixels),linewidth=3, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(matched_pixels,1)), np.transpose(matched_pixels[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 2)
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True)
return fig2
def plot_scatter_intensities(intensity_cols, experiment1, experiment2="Control_WCL", path=""):
"""
Create a scatter plot of the per-protein raw intensities between
two experiments. Experiment names should be the names of the columns
minus the "Intensity " part.
"""
if "Intensity %s"%experiment1 not in intensity_cols.columns:
raise ValueError("Supplied column name not in dataframe")
f = plt.figure(figsize=(3, 3))
plt.loglog()
tmp_prot_groups = intensity_cols.loc[:, ["Intensity %s"%experiment1, "Intensity %s"%experiment2]]
sns.regplot(tmp_prot_groups.ix[:, 1], tmp_prot_groups.ix[:, 0], fit_reg=False)
xs = np.linspace(tmp_prot_groups.min()[0], tmp_prot_groups.max()[0])
plt.plot(xs, tmp_prot_groups.sum()[0]/tmp_prot_groups.sum()[1]*xs, "r", alpha=0.4)
sns.axlabel("%s Raw Intensities"%experiment1.replace("_", " "),
"%s Raw Intensities"%experiment2.replace("_", " "))
sns.despine()
plt.tight_layout()
plt.savefig(os.path.join(path, "%sv%s_intensities.png"%(experiment1, experiment2)), dpi=300)
def plot_pca_components(self, fig1, gs, pca_components, required_pcs, plot_title='Habneula', gridspecs='[0,0]'):
ax1 = eval('fig1.add_subplot(gs' + gridspecs + ')')
with sns.axes_style('darkgrid'):
for ii in range(size(pca_components, 1)):
if ii in required_pcs:
plt.plot(pca_components[:, ii], '-', linewidth=5, label=str(ii))
else:
plt.plot(pca_components[:, ii], '--', label=str(ii))
plt.title(plot_title, fontsize=14)
sns.axlabel("Time (seconds)", "a.u")
plt.locator_params(axis='y', nbins=4)
sns.axlabel("Time (seconds)", "a.u")
ax1.legend(prop={'size': 14}, loc='center left', bbox_to_anchor=(1, 0.5), ncol=1, fancybox=True,
shadow=True)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
ax1.locator_params(axis='y', pad=50, nbins=2)
plt.ylim((min(pca_components) - 0.0001, max(pca_components) + 0.0001))
self.plot_vertical_lines_onset()
self.plot_vertical_lines_offset()
def universal_graph_func(text_variable,title_string,save_file_name_string):
sns.set(font_scale=.8)
sns.axlabel('Color', 'Frequency')
# colors from http://www.color-hex.com and wikipedia
flatui = ["#4b0082", "#ffd700", "#e6e6fa", "#ffff00", "#FF2400", "#ff6eb4", "#d2b48c", "#ff00ff", "#0000ff",
"#C8A2C8",
"#800080", "#FF007F", "#FD3F92", "#000000", "#dc143c", "#CCCCFF", "#ffffff", "#ff0000", "#631919",
"#fffff0",
"#ffa500", "#730000", "#808000", "#00ffff", "#c0c0c0", "#808080", "#7fffd4", "#808080", "#008000",
"#f5f5dc",
"#329999", "#f0ffff", "#FFEF00"]
custom_palette = sns.color_palette(flatui)
colors = text_variable[0]
occurences = text_variable[1]
ax = sns.barplot(colors, occurences, palette = custom_palette)
fig = ax.get_figure()
for item in ax.get_xticklabels():
item.set_rotation(45)
sns.plt.title(title_string)
fig.savefig(save_file_name_string)
fig.clf()
def plot_heatmap(df, title, file_name, vmax=1.5, vmin=-1.5, cmap=sns.diverging_palette(220, 20, n=7, as_cmap=True), figsize=(16, 8), dpi=300):
sns.set_style("white")
fig = plt.figure(figsize=figsize)
df = df.reindex(['G', 'A', 'V', 'I', 'L', 'M', 'F', 'Y', 'W', 'C', 'S', 'T', 'P', 'D', 'E', 'N', 'Q', 'H', 'K', 'R', '*'])
df.index.values[-1:] = ['STOP']
if 77 in df.columns:
df = df.loc[:, :76]
if 0 in df.columns:
df.drop(0, axis=1, inplace=True)
sns.heatmap(df, square=True, linewidths=0.25, xticklabels=2, cmap=cmap, vmin=vmin, vmax=vmax)
yeast_ubq = np.array(utils.canonical_yeast_ubq)
mask = df.apply(lambda x: x.index.isin(np.where(yeast_ubq == x.name)[0]+1), axis=1)
df.loc[:, :] = 1
sns.heatmap(df, mask=~mask, cmap=sns.dark_palette("grey", n_colors=1, as_cmap=True, reverse=True), xticklabels=2, cbar=False, square=True, alpha=0.6)
plt.yticks(rotation=0)
plt.title(title, fontsize=16)
pos = fig.axes[1].get_position()
fig.axes[1].set_position([pos.x0, 0.4, 0.1, 0.2])
sns.axlabel("Ub position", "Amino acid")
if file_name is not None:
plt.savefig(file_name, dpi=dpi, bbox_inches='tight', pad_inches=0)
return fig
def graph_func3(grimm):
sns.set( font_scale=.8)
sns.axlabel('Color', 'Frequency' )
#colors from http://www.color-hex.com and wikipedia
#flatui = ["#521515", "#fffff0", "#4b0082","#ffd700", "#7fffd4", "#e6e6fa", "#ffff00",
# "#dc143c", "#ffc0cb", "#660000", "#808000", "#00ffff", "#d2b48c", "#c0c0c0",
# "#ff00ff", "#0000ff", "#808080", "#ffec8b","#ab82ff", "#ee4000", "#993299",
# "#ffaeb9", "#FF00FF", "#808080", '#f0ffff', "#008000", "#f5f5dc", "#008080",
# "#CCCCFF","#ffa500", "#000000", "#ffffff", "#ff0000"]
flatui = ["#4b0082","#ffd700", "#e6e6fa", "#ffff00", "#FF2400", "#ff6eb4", "#d2b48c", "#ff00ff", "#0000ff", "#C8A2C8",
"#800080", "#FF007F", "#FD3F92", "#000000", "#dc143c", "#CCCCFF", "#ffffff", "#ff0000", "#631919", "#fffff0",
"#ffa500","#730000", "#808000","#00ffff","#c0c0c0","#808080", "#7fffd4", "#808080", "#008000", "#f5f5dc",
"#329999", "#f0ffff", "#FFEF00"]
custom_palette = sns.color_palette(flatui)
colors = grimm[0]
occurences = grimm[1]
ax3 = sns.barplot(colors, occurences, palette = custom_palette)
fig3 = ax3.get_figure()
for item in ax3.get_xticklabels():
item.set_rotation(45)
sns.plt.title('Color Word Frequencies in Brothers Grimm Stories')
fig3.savefig('grimm_chart.png')
fig3.clf()
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
data = np.array(list(map(lambda x: list(map(float, x.split("\t"))),
"""10 100 250 750 2250 6200 10000 12000
0.244 10.666 20.033 45.213 195.22 949.797 1413.038 1836.463
0.001 0.008 0.059 1.201 29.594 595.976 2522.441 8134.59""".split("\n"))))
sns.set(style="white", palette="muted")
b, g, r, p = sns.color_palette("muted", 4)
ax = sns.tsplot(data[1], time=data[0], condition="Learning",color=g)
plt.tight_layout()
ax.fill_between(data[0], 1e-12, data[1], color=g, alpha=0.25)
ax2 = sns.tsplot(data[1] + data[2],condition="Learning + Inversion", time=data[0], color=b)
ax2.fill_between(data[0], data[1] + data[2], data[1], color=b, alpha=0.25)
plt.legend(loc='upper left')
plt.title("System Computation time vs. Recommendable Documents")
sns.axlabel("# Of Documents", "Time (s)")
plt.savefig("images/computation_time.pdf", bbox_inches="tight")
#
# - Plotting function parameters, e.g. sns.distplot(kde=False)
# - Seaborn functions, called on the sns Seaborn object synonym
#
# Similar to how you use plt, matplotlib's import synonym, to customize matplotlib graphics, you use sns, seaborn's import synonym, to refer to a plot and call functions from <a href = "https://stanford.edu/~mwaskom/software/seaborn/api.html#style-frontend">Seaborn's API</a> to customize it. For example, to set the x-axis and y-axis labels, use the <a href = "https://stanford.edu/~mwaskom/software/seaborn/generated/seaborn.axlabel.html#seaborn.axlabel">.axlabel() function</a>:
#
# sns.distplot(births['prglngth'], kde=False)
# sns.axlabel('Pregnancy Length, weeks', 'Frequency')
# In[5]:
import seaborn as sns
get_ipython().magic('matplotlib inline')
sns.distplot(births['prglngth'], kde=False)
sns.axlabel('Pregnancy Length, weeks', 'Frequency')
# ###6: Practice: customizing distplot()
# Now it's your turn to practice customizing histograms using Seaborn!
# ####Instructions
# Plot a histogram of the birthord column with the following tweaks:
#
# x-axis label: Birth number
# y-axis label: Frequency
# style: "dark"
# In[6]:
def plot_pca_figures(pca, maps, pts, clrs, recon,tt,unique_clrs, matched_pixels,matched_signals, Exp_Folder, filename_suffix, Exp_Name):
#Plotting as pdf
Figure_PDFDirectory = Exp_Folder+'Figures'+filesep
if not os.path.exists(Figure_PDFDirectory):
os.makedirs(Figure_PDFDirectory)
pp = PdfPages(Figure_PDFDirectory+filename_suffix+'_PCA.pdf')
sns.set_context("poster")
dIPN = [Exp_Name.index(s) for s in Exp_Name if "dIPN" in s]
vIPN = [Exp_Name.index(s) for s in Exp_Name if "vIPN" in s]
############ Plot Colormaps of scores ############
#If there is only one stack, else plot each stack
if len(maps.shape)==3:
#Plot colored maps for each stack
with sns.axes_style("white"):
fig2 = plt.imshow(maps[:,:,:].transpose((1,0,2)))
plt.title(Exp_Name[0])
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
else:
#Plot dIPN
fig2 = plt.figure()
for ii in range(0,np.size(dIPN,0)):
with sns.axes_style("white"):
plt.subplot(2,3,ii+1)
plt.imshow(maps[:,:,dIPN[ii],:])
plt.axis('off')
plt.title(Exp_Name[dIPN[ii]][15:18],fontsize=10)
# plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
#Plot vIPN
fig2 = plt.figure()
for ii in range(0,np.size(vIPN,0)):
with sns.axes_style("white"):
plt.subplot(2,3,ii+1)
plt.imshow(maps[:,:,vIPN[ii],:])
plt.axis('off')
plt.title(Exp_Name[vIPN[ii]][15:18], fontsize=10)
# plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
########### Plot components ##################
fig2 = plt.figure()
sns.set_context("talk", font_scale=1.25)
with sns.axes_style("dark"):
ax1 = plt.subplot(231)
plt.plot(pca.comps.T);
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
A = []
for ii in xrange(0,np.size(pca.comps.T, 0)):
A = np.append(A, ['comp' + str(ii+1)])
ax1.legend(A, loc=4)
plot_vertical_lines()
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
#Plot mean signals according to color and boxplot of number of pixels in each plane
with sns.axes_style("dark"):
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.subplot(234)
sns.tsplot(np.array(matched_signals[ii].clr_grped_signal), ci=95, err_style="ci_band", color=unique_clrs[ii])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
plot_vertical_lines()
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
matching = [Exp_Name.index(s) for s in Exp_Name if "vIPN" in s]
temp_matched_pixels = matched_pixels[:,matching]
with sns.axes_style("white"):
fig2 = plt.subplot(232)
fig2 = sns.boxplot(np.transpose(temp_matched_pixels),linewidth=2, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(temp_matched_pixels,1)), np.transpose(temp_matched_pixels[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 4)
plt.title('vIPN')
# plt.ylim((0, 1000))
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True);
matching = [Exp_Name.index(s) for s in Exp_Name if "dIPN" in s]
temp_matched_pixels = matched_pixels[:,matching]
with sns.axes_style("white"):
fig2 = plt.subplot(233)
fig2 = sns.boxplot(np.transpose(temp_matched_pixels),linewidth=2, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(temp_matched_pixels,1)), np.transpose(temp_matched_pixels[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 4)
# plt.ylim((0, 1000))
sns.axlabel("Colors", "Number of Pixels")
plt.title('dIPN')
sns.despine(offset=10, trim=True);
#Create an lm plot seperating Before and After number of pixels
#Make Panda data frame
A = np.zeros((3,np.size(matched_pixels,0)*np.size(matched_pixels,1)), dtype=np.int)
count = 0
for ii in range(0,np.size(matched_pixels,0)):
for jj in range(0,np.size(matched_pixels,1)):
A[0,count] = matched_pixels[ii,jj]
A[1,count] = ii
if jj in [0,1,3,5]:
A[2,count] = 1 #vIPN
else:
A[2,count] = 0 #dIPN
count = count+1
A = np.transpose(A)
B = pd.DataFrame({'Pixel':A[:,0], 'response':A[:,1], 'IPN':A[:,2]})
B["Region"] = B.IPN.map({1: "vIPN", 0: "dIPN"})
#Plot mean projection
with sns.axes_style("white"):
fig2 = plt.subplot(235)
matching = [Exp_Name.index(s) for s in Exp_Name if "vIPN" in s]
temp = np.max(maps[:,:,matching,:], axis=2)
plt.imshow(temp.astype(np.float16).transpose((1,0,2)))
plt.axis('off')
plt.title('Max vIPN')
fig2 = plt.subplot(236)
matching = [Exp_Name.index(s) for s in Exp_Name if "dIPN" in s]
temp = np.max(maps[:,:,matching,:], axis=2)
plt.imshow(temp.astype(np.float16).transpose((1,0,2)))
plt.axis('off')
plt.title('Max dIPN')
plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
with sns.axes_style("dark"):
fig3 = plt.figure()
g = sns.lmplot("Pixel", "IPN", B, y_jitter=.20, hue="response", fit_reg=False, palette=unique_clrs, markers="s", scatter_kws={"s": 50})
g.set(ylim = (-0.2, 1.2), yticks=[0, 1], yticklabels=["dIPN", "vIPN"])
plt.axhline(y=0.5, linestyle='-', color='w', linewidth=0.5)
fig3 = plt.gcf()
pp.savefig(fig3)
plt.close()
with sns.axes_style("dark"):
fig3 = plt.figure()
g = sns.lmplot("Pixel", "IPN", B, y_jitter=.20, fit_reg=True, palette=unique_clrs, markers="s", scatter_kws={"s": 50})
g.set(ylim = (-0.2, 1.2), yticks=[0, 1], yticklabels=["dIPN", "vIPN"])
plt.axhline(y=0.5, linestyle='-', color='w', linewidth=0.5)
fig3 = plt.gcf()
pp.savefig(fig3)
plt.close()
pp.close()
def plot_pca_maps_for_stacks(pca, maps, pts, clrs, recon, unique_clrs, matched_pixels, matched_signals, num_z_planes,\
Exp_Folder, filename_save_prefix, Stimulus_Name, stim_start, stim_end):
filesep = os.path.sep
# To save as pdf create file
Figure_PDFDirectory = Exp_Folder+filesep+'Figures'+filesep
if not os.path.exists(Figure_PDFDirectory):
os.makedirs(Figure_PDFDirectory)
pp = PdfPages(Figure_PDFDirectory+filename_save_prefix+'_PCA_Stacks.pdf')
sns.set_context("poster")
if num_z_planes!=0:
Filenames_stim = [ii + ' Z='+ str(jj) for jj in num_z_planes for ii in Stimulus_Name]
else:
Filenames_stim = [ii + ' Z='+ str(jj+1) for jj in range(0,np.size(maps,2)/np.size(Stimulus_Name)) for ii in Stimulus_Name]
############ Plot Colormaps of scores ############
for ii in range(0,np.size(Filenames_stim)):
with sns.axes_style("white"):
fig2 = plt.imshow(maps[:,:,ii,:])
plt.title((Filenames_stim[ii]))
plt.axis('off')
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
########### Plot components ##################
fig2 = plt.figure()
sns.set_context("talk", font_scale=1.25)
with sns.axes_style("darkgrid"):
ax1 = plt.subplot(221)
plt.plot(pca.comps.T);
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
A = []
for ii in xrange(0,np.size(pca.comps.T, 0)):
A = np.append(A, [str(ii+1)])
ax1.legend(A, loc=4)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
plot_vertical_lines(stim_start,stim_end)
#Plot mean signals according to color and boxplot of number of pixels in each plane
with sns.axes_style("darkgrid"):
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.subplot(223)
sns.tsplot(np.array(matched_signals[ii].clr_grped_signal), linewidth=3, ci=95, err_style="ci_band", color=unique_clrs[ii])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
plot_vertical_lines(stim_start,stim_end)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
with sns.axes_style("white"):
fig2 = plt.subplot(222)
fig2 = sns.boxplot(np.transpose(matched_pixels),linewidth=3, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(matched_pixels,1)), np.transpose(matched_pixels[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True);
plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
for stim in xrange(0,np.size(Stimulus_Name)):
with sns.axes_style("white"):
same_stim_folders = [Filenames_stim.index(ii) for ii in Filenames_stim if ii.find(Stimulus_Name[stim])==0]
fig2 = plt.subplot(221)
matched_pixels_stim = matched_pixels[:,same_stim_folders]
if np.size(matched_pixels_stim,1) == 1:
for ii in xrange(0,np.size(matched_pixels_stim,0)):
fig2 = plt.plot(ii+1,np.transpose(matched_pixels_stim[ii,:]),'o', color=unique_clrs[ii])
plt.xlim([0,np.size(matched_pixels_stim,0)+1])
else:
fig2 = sns.boxplot(np.transpose(matched_pixels_stim),linewidth=3, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(matched_pixels_stim,1)), np.transpose(matched_pixels_stim[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.subplot(223)
sns.tsplot(np.array(matched_signals[ii].clr_grped_signal), linewidth=3, ci=95, err_style="ci_band", color=unique_clrs[ii])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
plot_vertical_lines(stim_start,stim_end)
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
fig2 = plt.subplot(222)
temp = (np.mean(maps[:,:,same_stim_folders,:], axis=2))
plt.imshow(temp.astype(np.float16))
plt.axis('off')
plt.title('Mean projection' + Stimulus_Name[stim])
plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
pp.close()
from matplotlib import pyplot as plt
import seaborn as sns
import pandas as pd
import numpy as np
# Script used to generate timing plot for different error correction methods
timings = np.array([12, 15, 3, 3, 10, 67, 62])
IDs = ["Quake", "Bless", "Musket", "BFC", "Seecer", "BayesHammer", "MultiRes"]
sns.barplot(x=IDs, y=timings)
sns.axlabel("Algorithms", "Timings(in minutes)")
sns.plt.title("Comparison of timings for error correction algorithms")
plt.figure(figsize=(16, 10))
sns.set_context("poster", font_scale=0.6) #rc={"lines.linewidth": 2.5}
""" # the view of everthing
sns.heatmap(refpubyear_count, linewidths=.5, yticklabels=3)
ax.xaxis.tick_top()
ax.invert_yaxis()
sns.axlabel(conf + ' paper-published-year', 'reference-published-year')
"""
# the subset of reference years
ax = sns.heatmap(refpubyear_count.loc[1975:2015], linewidths=.5, annot=True, fmt=".0f", cmap="GnBu")
ax.xaxis.tick_top()
ax.invert_yaxis()
sns.axlabel(conf + ' paper-published-year', 'reference-published-year')
plt.savefig(os.path.join(plot_dir, conf, conf+'_year_ref.png'), transparent=True)
""" # box plot of the age of paper being cited
dataframe: df_citing
y: RefPubYear
y: ref age (paperpubyear - refpubyear)
x (group by): PaperPubYear
"""
plt.figure(figsize=(16, 4))
df_citing['RefAge'] = - df_citing['PaperPubYear'] + df_citing['RefPubYear']
axb = sns.boxplot(x="PaperPubYear", y="RefAge", data=df_citing,
linewidth=1., palette=bar_colrs3)
ylim = axb.get_ylim()
axb.set_ylim([max(-60, ylim[0]), min(5, ylim[1])])
header=None,
names=["inst", "time"],
usecols=["time"],
na_values=["t"])
except IOError:
next
color = '#000000'
format = 'pdf'
if sys.argv[1] == "eps":
color = '#796045'
format = 'eps'
sns.set(font_scale=1.41)
sns.set_style("whitegrid",
rc={
"grid.color": color,
"xtick.color": color,
"ytick.color": color,
"axes.edgecolor": color,
"axes.labelcolor": color,
"text.facecolor": 'white',
})
sns.set_palette("Set1", desat=0.22)
sns.axlabel("time needed", "") #"heuristic")
ax = sns.boxplot(frame, orient="h", width=0.79)
ax.get_figure().subplots_adjust(left=0.24)
ax.set(xscale="log")
ax.get_figure().savefig('build/tsplib_med_time.' + format,
format=format,
transparent=True)
# Simple plot of linear, quadratic, and cubic curves
x = np.linspace(0, 2, 100)
plt.plot(x, x, label='linear')
plt.plot(x, x**2, label='quadratic')
plt.plot(x, x**3, label='cubic')
plt.xlabel('x label')
plt.ylabel('y label')
plt.title("Simple Plot")
plt.legend(loc="best")
plt.show()
# Histogram
x = np.random.normal(size=1000)
sns.distplot(x, bins=20, kde=True, rug=False, label="Histogram w/o Density")
sns.axlabel("Value", "Frequency")
plt.title("Histogram of a Random Sample from a Normal Distribution")
plt.legend()
plt.show()
# Scatter plot
mean, cov = [5, 10], [(1, .5), (.5, 1)]
data = np.random.multivariate_normal(mean, cov, 200)
data_frame = pd.DataFrame(data, columns=["x", "y"])
sns.jointplot(x="x", y="y", data=data_frame, kind="reg").set_axis_labels("x", "y")
plt.suptitle("Joint Plot of Two Variables with Bivariate and Univariate Graphs")
plt.show()
# Pairwise bivariate
for s in range(100):
U[s, :] = e.policy_batch.eval(e.sess, X).flat
tshist(axarr[i, 1], X, U)
# column 2 log-loss surfaces after learning
x0v, u0v = np.meshgrid(np.linspace(0, 1, 30), np.linspace(-1, 1, 30))
x0v, u0v = x0v.reshape((-1, 1)), u0v.reshape((-1, 1))
x1v = np.array([sim.fstep(x, u0v[j]) for j, x in enumerate(x0v)]).reshape(
(-1, 1))
Lr = e.eval_1d(e.loss_log, x0v, u0v, x1v)
sns.heatmap(Lr.reshape(30, 30),
ax=axarr[i, 2],
cmap='gray',
vmin=-13,
vmax=-.5)
sns.axlabel('X', 'A')
# column 3 - prediction loss under a variety of
#Lp = e.eval_1d(e.loss_abs_p,x0v,u0v,x1v)
#sns.heatmap(Lp.reshape(30,30),ax=axarr[i,3],cmap='gray',vmin=0.,vmax=1.)
#Xr = e.eval_1d(e.loss_predict,X,X,X) # the second 2 args are hacks
#axarr[i,3].plot(X,Xr)
print(c)
# # test
# x0v = X
# u0v = np.ones((e.batch_size,1))
# x1v = np.ones((e.batch_size,1)) # irrelevant
# xp = e.eval_1d(e.x_predict,x0v,u0v,x1v)
# pdb.set_trace()
print ('chisq test statistic = {}; p-value = {}'.format(*chisq))
# graph plotting with seaborn
sns.set_style('white')
sns.set_style('ticks')
plt.figure(figsize=(8,5))
sns.distplot(obs_distances,
bins=np.linspace(0, WINDOW, 41), norm_hist=True, color='#d8b365',
hist_kws={'alpha': 0.3}, kde_kws={'bw': increment}, label='Observed')
sns.distplot(exp_distances,
bins=np.linspace(0, WINDOW, 41), norm_hist=True, color='#5ab4ac',
hist_kws={'alpha': 0.3}, kde_kws={'bw': increment}, label='Expected')
if args.dist == 'upstream_to_start':
sns.plt.xlim(int(WINDOW * 1.01), int(WINDOW * -0.01))
sns.axlabel("Distance to gene 5' end (bp)", 'Density')
elif args.dist == 'downstream_from_start':
sns.plt.xlim(int(WINDOW * -0.01), int(WINDOW * 1.01))
sns.axlabel("Distance from gene 5' end (bp)", 'Density')
elif args.dist == 'upstream_to_end':
sns.plt.xlim(int(WINDOW * 1.01), int(WINDOW * -0.01))
sns.axlabel("Distance to gene 3' end (bp)", 'Density')
elif args.dist == 'downstream_from_end':
sns.plt.xlim(int(WINDOW * -0.01), int(WINDOW * 1.01))
sns.axlabel("Distance from gene 3' end (bp)", 'Density')
plt.legend(loc=9, ncol=2)
sns.despine(offset=10, trim=True)
# save plot
fig = plt.gcf()
plt.show()
# Look at the distribution of quality by wine type
red_wine = wine.ix[wine['type']=='red', 'quality']
white_wine = wine.ix[wine['type']=='white', 'quality']
sns.set_style("dark")
print("""\nsns.distplot(red_wine,
norm_hist=True, kde=False, color=red, label=Red wine)""")
print(sns.distplot(red_wine, \
norm_hist=True, kde=False, color="red", label="Red wine"))
print("""\nsns.distplot(white_wine,
norm_hist=True, kde=False, color=white, label=White wine)""")
print(sns.distplot(white_wine, \
norm_hist=True, kde=False, color="white", label="White wine"))
sns.axlabel("Quality Score", "Density")
plt.title("Distribution of Quality by Wine Type")
plt.legend()
#plt.show()
# Test whether mean quality is different between red and white wines
print("\nwine.groupby(['type'])[['quality']].agg(['std', 'mean'])")
print(wine.groupby(['type'])[['quality']].agg(['std', 'mean']))
tstat, pvalue, df = sm.stats.ttest_ind(red_wine, white_wine)
print("\n'tstat: %.3f pvalue: %.4f' % (tstat, pvalue)")
print('tstat: %.3f pvalue: %.4f' % (tstat, pvalue))
# Fit a multivariate linear regression model
#wine_standardized = (wine - wine.mean()) / wine.std()
#formula_all = 'quality ~ alcohol + chlorides + citric_acid + density + fixed_acidity + free_sulfur_dioxide + pH + residual_sugar + sulphates + total_sulfur_dioxide + volatile_acidity'
my_formula = 'quality ~ alcohol + chlorides + citric_acid + density + fixed_acidity + free_sulfur_dioxide + pH + residual_sugar + sulphates + total_sulfur_dioxide + volatile_acidity'
if args.end:
endTime = re.sub('\.{1}','T',args.end)
endTime+=":00"
firstIndex = dateRange.index(endTime)
else:
firstIndex = 0
palette = it.cycle(sea.color_palette())
fig = plt.figure()
sea.set_style('darkgrid')
fig.suptitle('MS Band Hourly Summary',fontsize=16)
ax1 = fig.add_subplot(311)
ax1.plot_date(x[firstIndex:lastIndex],caloriesBurned[firstIndex:lastIndex],color=next(palette),linestyle='-',fillstyle='none')
sea.axlabel('','Calories Burned')
ax2 = fig.add_subplot(312) # 3 rows, 1 column, plot #2
ax2.plot_date(x[firstIndex:lastIndex],stepsTaken[firstIndex:lastIndex],color=next(palette),linestyle='-',fillstyle='none')
sea.axlabel('','Steps Taken')
ax3 = fig.add_subplot(313) # 3 rows, 1 column, plot #3
ax3.plot_date(x[firstIndex:lastIndex],avgHeartRate[firstIndex:lastIndex],color=next(palette),linestyle='-',fillstyle='none')
ax3.plot_date(x[firstIndex:lastIndex],lowHeartRate[firstIndex:lastIndex],color=next(palette),linestyle='-',fillstyle='none')
ax3.plot_date(x[firstIndex:lastIndex],peakHeartRate[firstIndex:lastIndex],color=next(palette),linestyle='-',fillstyle='none')
ax3.xaxis.set_major_formatter(dates.DateFormatter('%m/%d/%Y %H:%M'))
fig.autofmt_xdate() # angle the dates for easier reading
sea.axlabel('Date','Heart Rate')
import seaborn as sns
import matplotlib.pyplot as plt
from datos import data
sns.set(style="whitegrid")
d = data('mtcars')
sns.countplot(x="gear", hue="cyl", data=d)
sns.axlabel("Number of Gears", "Frequency")
plt.title('Car Distribution by Gear and Cylindres', family='Serif', size=16)
plt.show()
roi_names, roi_coords = load_msdl_names_and_coords()
stat_av = read_test('pc', 'av', 'avg')
stat_v = read_test('pc', 'v', 'avg')
stat2 = read_test2('pc', ['av', 'v'], 'avg')
i, j = np.unravel_index(stat2.argmax(), stat2.shape)
print 'av 1sample pval :', stat_av[i, j]
print 'v 1sample pval :', stat_v[i, j]
print roi_names[i], roi_names[j]
print i, j
m = np.eye(2)
m[1,0] = stat2[i, j]
m[0,1] = m[1,0]
plot_connectome(m, [roi_coords[i], roi_coords[j]])
conn = []
behav = []
for i in range(len(dataset.subjects)):
c = load_dynacomp_fc(dataset.subjects[i], session='func1', metric='pc', msdl=True)
conn.append(c[i, j])
b = behav_data[i]['postRT'] - behav_data[i]['preRT']
behav.append(b)
sns.jointplot(np.array(conn), np.array(behav), kind='kde')
sns.axlabel('Connectivity', 'Behavior')
def plot_regression_error(clf, X_train, y_train, name):
predicted = cross_val_predict(clf, X_train, y_train, cv=5)
sns.regplot(x=y_train, y=predicted)
sns.axlabel("actual", "predicted")
plt.savefig("plot_validation_" + name + ".png")
def plot_pca_figures(pca, maps, pts, clrs, recon,tt,unique_clrs, matched_pixels,matched_signals, Exp_Folder, filename_suffix, Exp_Name):
#Plotting as pdf
Figure_PDFDirectory = Exp_Folder+'Figures'+filesep
if not os.path.exists(Figure_PDFDirectory):
os.makedirs(Figure_PDFDirectory)
pp = PdfPages(Figure_PDFDirectory+filename_suffix+'_PCA.pdf')
sns.set_context("poster")
#Pick experiemnt names and seperate to indeces for plotting
Fishnum = np.unique([np.int(ii[15:18]) for ii in Exp_Name])
############ Plot Colormaps of scores ############
#If there is only one stack, else plot each stack
if len(maps.shape)==3:
#Plot colored maps for each stack
with sns.axes_style("white"):
fig2 = plt.imshow(maps[:,:,:].transpose((1,0,2)))
plt.title(Exp_Name[0])
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
else:
count = 0
for ii in range(0,np.size(Fishnum,0)):
fig2 = plt.figure()
count2=3
with sns.axes_style("white"):
for jj in range(3):
print str(ii) + ' ' + str(jj)
if ii == 0 and jj == 2:
continue
if "Before" in Exp_Name[count]:
plt.subplot(2,2,1)
plt.imshow(maps[:,:,count,:].transpose((1,0,2)))
plt.axis('off')
plt.title(Exp_Name[count][15:-4])
count = count+1
else:
plt.subplot(2,2,count2)
plt.imshow(maps[:,:,count,:].transpose((1,0,2)))
plt.axis('off')
plt.title(Exp_Name[count][15:-4])
count = count+1
count2=count2+1
plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
########### Plot components ##################
fig2 = plt.figure()
sns.set_context("talk", font_scale=1.25)
with sns.axes_style("dark"):
ax1 = plt.subplot(231)
plt.plot(pca.comps.T);
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
A = []
for ii in xrange(0,np.size(pca.comps.T, 0)):
A = np.append(A, ['comp' + str(ii+1)])
ax1.legend(A, loc=4)
plot_vertical_lines()
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
#Plot mean signals according to color and boxplot of number of pixels in each plane
with sns.axes_style("dark"):
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.subplot(234)
sns.tsplot(np.array(matched_signals[ii].clr_grped_signal), ci=95, err_style="ci_band", color=unique_clrs[ii])
plt.locator_params(axis = 'y', nbins = 4)
sns.axlabel("Time (seconds)","a.u")
plot_vertical_lines()
plt.axhline(y=0, linestyle='-', color='k', linewidth=1)
matching = [Exp_Name.index(s) for s in Exp_Name if "Before" in s]
temp_matched_pixels = matched_pixels[:,matching]
with sns.axes_style("white"):
fig2 = plt.subplot(232)
fig2 = sns.boxplot(np.transpose(temp_matched_pixels),linewidth=2, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(temp_matched_pixels,1)), np.transpose(temp_matched_pixels[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 4)
# plt.ylim((0, 600))
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True);
matching = [Exp_Name.index(s) for s in Exp_Name if "After" in s]
temp_matched_pixels = matched_pixels[:,matching]
with sns.axes_style("white"):
fig2 = plt.subplot(233)
fig2 = sns.boxplot(np.transpose(temp_matched_pixels),linewidth=2, widths=.5, color=unique_clrs)
for ii in range(0,np.size(unique_clrs,0)):
fig2 = plt.plot(np.repeat(ii+1,np.size(temp_matched_pixels,1)), np.transpose(temp_matched_pixels[ii,:]),'s', \
color=unique_clrs[ii], markersize=5, markeredgecolor='k', markeredgewidth=2)
plt.locator_params(axis = 'y', nbins = 4)
# plt.ylim((0, 600))
sns.axlabel("Colors", "Number of Pixels")
sns.despine(offset=10, trim=True);
#Create an lm plot seperating Before and After number of pixels
#Make Panda data frame
A = np.zeros((3,np.size(matched_pixels,0)*np.size(matched_pixels,1)), dtype=np.int)
count = 0
for ii in range(0,np.size(matched_pixels,0)):
for jj in range(0,np.size(matched_pixels,1)):
A[0,count] = matched_pixels[ii,jj]
A[1,count] = ii
if jj in [0,2,5]:
A[2,count] = 1 #Before Lesion
else:
A[2,count] = 0 #After Lesion
count = count+1
A = np.transpose(A)
B = pd.DataFrame({'Pixel':A[:,0], 'response':A[:,1], 'Lesion':A[:,2]})
B["BeforeAfter"] = B.Lesion.map({1: "Before", 0: "After"})
#Plot mean projection
with sns.axes_style("white"):
fig2 = plt.subplot(438)
matching = [Exp_Name.index(s) for s in Exp_Name if "Before" in s]
temp = np.max(maps[:,:,matching,:], axis=2)
plt.imshow(temp.astype(np.float16).transpose((1,0,2)))
plt.axis('off')
plt.title('Max DRBefore')
fig2 = plt.subplot(4,3,11)
matching = [Exp_Name.index(s) for s in Exp_Name if "After" in s]
temp = np.max(maps[:,:,matching[1:],:], axis=2)+0.2
plt.imshow(temp.astype(np.float16).transpose((1,0,2)))
plt.axis('off')
plt.title('Max DRAfter')
plt.tight_layout()
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
with sns.axes_style("dark"):
fig3 = plt.figure()
g = sns.lmplot("Pixel", "Lesion", B, y_jitter=.20, hue="response", fit_reg=False, palette=unique_clrs, markers="s", scatter_kws={"s": 50})
g.set(ylim = (-0.2, 1.2), yticks=[0, 1], yticklabels=["After", "Before"])
plt.axhline(y=0.5, linestyle='-', color='w', linewidth=0.5)
fig3 = plt.gcf()
pp.savefig(fig3)
plt.close()
with sns.axes_style("dark"):
fig3 = plt.figure()
g = sns.lmplot("Pixel", "Lesion", B, y_jitter=.20, fit_reg=True, palette=unique_clrs, markers="s", scatter_kws={"s": 50})
g.set(ylim = (-0.2, 1.2), yticks=[0, 1], yticklabels=["After", "Before"])
plt.axhline(y=0.5, linestyle='-', color='w', linewidth=0.5)
fig3 = plt.gcf()
pp.savefig(fig3)
plt.close()
matched_pixels1 = np.delete(matched_pixels, [0],0)
A = np.zeros((3,np.size(matched_pixels1,0)*np.size(matched_pixels1,1)), dtype=np.float)
count = 0
for ii in range(0,np.size(matched_pixels1,0)):
for jj in range(0,np.size(matched_pixels1,1)):
A[0,count] = matched_pixels1[ii,jj]/220
if ii == 2 or ii == 3 or ii == 4:
A[1,count] = 1
else:
A[1,count] = 0
if jj in [0,2,5]:
A[2,count] = 0 #Before Lesion
else:
A[2,count] = 1 #After Lesion
count = count+1
A = np.transpose(A)
B = pd.DataFrame({'Cells':A[:,0], 'response':A[:,1], 'Lesion':A[:,2]})
B["Bef_aft"] = B.Lesion.map({0: "Before", 1: "CAfter"})
B["Response"] = B.response.map({0: "Excitatory On", 1: "Inhibitory On"})
fig2 = plt.figure()
fig2 = sns.factorplot("Response", "Cells", "Bef_aft", B, kind="point", dodge=0.05)
fig2 = plt.gcf()
pp.savefig(fig2)
plt.close()
pp.close()
# - Plotting function parameters, e.g. sns.distplot(kde=False)
# - Seaborn functions, called on the sns Seaborn object synonym
#
# Similar to how you use plt, matplotlib's import synonym, to customize matplotlib graphics, you use sns, seaborn's import synonym, to refer to a plot and call functions from <a href = "https://stanford.edu/~mwaskom/software/seaborn/api.html#style-frontend">Seaborn's API</a> to customize it. For example, to set the x-axis and y-axis labels, use the <a href = "https://stanford.edu/~mwaskom/software/seaborn/generated/seaborn.axlabel.html#seaborn.axlabel">.axlabel() function</a>:
#
# sns.distplot(births['prglngth'], kde=False)
# sns.axlabel('Pregnancy Length, weeks', 'Frequency')
# In[5]:
import seaborn as sns
get_ipython().magic('matplotlib inline')
sns.distplot(births['prglngth'], kde=False)
sns.axlabel('Pregnancy Length, weeks', 'Frequency')
# ###6: Practice: customizing distplot()
# Now it's your turn to practice customizing histograms using Seaborn!
# ####Instructions
# Plot a histogram of the birthord column with the following tweaks:
#
# x-axis label: Birth number
# y-axis label: Frequency
# style: "dark"
# In[6]:
