def make_autos_losses_plot(
figure, gs=None, pddata=None, axes=None, legend=True, labels=True):
if not axes:
axes = figure.add_subplot(gs[0, 0])
loss_mins = pddata.groupby("make")["losses"].min().values
loss_means = pddata.groupby("make")["losses"].mean().values
loss_maxs = pddata.groupby("make")["losses"].max().values
make_ids = geradados.get_make_ids(pddata)
min_color = colors[0]
mean_color = colors[3]
max_color = colors[2]
axes.set_title("Inverse Losses", fontsize=14)
mins_bar = axes.bar(make_ids, loss_mins, width=0.5, align="center",
color=min_color, alpha=0.7)
means_bar = axes.bar(make_ids, loss_means, width=0.5, align="center",
bottom=loss_mins, color=mean_color, alpha=0.7)
maxs_bar = axes.bar(make_ids, loss_maxs, width=0.5, align="center",
bottom=loss_means + loss_mins, color=max_color, alpha=0.7)
if labels:
axes.set_xticks(range(0, 13))
axes.set_xticklabels(geradados.get_make_labels(pddata))
axes.set_xlabel("Make", fontsize=12)
axes.set_ylabel("Inverse Losses", fontsize=12)
else:
axes.set_xticklabels([])
axes.set_yticklabels([])
if legend:
axes.legend([mins_bar, means_bar, maxs_bar], ["Min", "Mean", "Max"], loc=2)
if gs:
gs.tight_layout(figure)
return axes
def make_autos_loss_and_risk_plot(figure,
gs=None,
pddata=None,
axes=None,
x_label=True,
rotate_ticks=False):
if not axes:
axes = figure.add_subplot(gs[0, 0])
if rotate_ticks:
plt.xticks(rotation=70)
risk_mins = pddata.groupby("make")["riskiness"].min().values
risk_means = pddata.groupby("make")["riskiness"].mean().values
risk_maxs = pddata.groupby("make")["riskiness"].max().values
loss_mins = pddata.groupby("make")["losses"].min().values
loss_means = pddata.groupby("make")["losses"].mean().values
loss_maxs = pddata.groupby("make")["losses"].max().values
mins = risk_mins + loss_mins
means = risk_means + loss_means
maxs = risk_maxs + loss_maxs
make_ids = geradados.get_make_ids(pddata)
min_color = colors[0]
mean_color = colors[3]
max_color = colors[2]
axes.set_title(("Combined Losses and Riskiness Data\n"
"(Inverted, Normalized)"),
fontsize=16)
mins_bar = axes.bar(make_ids,
mins,
align="center",
color=min_color,
alpha=0.7)
means_bar = axes.bar(make_ids,
means,
align="center",
bottom=mins,
color=mean_color,
alpha=0.7)
maxs_bar = axes.bar(make_ids,
maxs,
align="center",
bottom=means + mins,
color=max_color,
alpha=0.7)
axes.set_xticks(range(0, 13))
axes.set_xticklabels(geradados.get_make_labels(pddata))
if x_label:
axes.set_xlabel("Make", fontsize=12)
axes.set_ylabel("Risco", fontsize=12)
axes.legend([mins_bar, means_bar, maxs_bar], ["Min", "Mean", "Max"], loc=2)
if gs:
gs.tight_layout(figure)
return axes
def make_autos_mpg_plot(figure, gs = None, pddata = None, axes = None):
if not axes:
axes = figure.add_subplot(gs[0, 0])
data = geradados.get_numeric_data(pddata)
axes.set_title("Ranges of Cidades e Velocidade", fontsize = 20)
axes.scatter(data["make"], data["highway mpg"], c = colors[3], s = 500, alpha = 0.4)
axes.scatter(data["make"], data["city mpg"], c = colors[0], s = 500, alpha = 0.4)
axes.set_xticks(range(0, 13))
axes.set_xticklabels(geradados.get_make_labels(pddata))
axes.set_xlabel("Make", fontsize = 16)
axes.set_ylabel("MPG", fontsize = 16)
city_patch = mpl.patches.Patch(color = colors[0], alpha = 0.7, label = "City")
highway_patch = mpl.patches.Patch(color=colors[3], alpha=0.7, label = "Highway")
axes.legend(handles=[city_patch, highway_patch], loc = 2)
if gs:
gs.tight_layout(figure)
return axes
def make_autos_price_plot(figure, gs=None, pddata=None, axes=None):
if not axes:
axes = figure.add_subplot(gs[0, 0])
min_data = pddata.groupby("make", sort=True)["price"].min()
max_data = pddata.groupby("make", sort=True)["price"].max()
mean_data = pddata.groupby("make", sort=True)["price"].mean()
make_ids = geradados.get_make_ids(pddata)
axes.set_title("Auto Price Ranges", fontsize=20)
axes.plot(make_ids, min_data, c=colors[2], linewidth=4, alpha=0.7)
axes.plot(make_ids, mean_data, c=colors[3], linewidth=4, alpha=0.7)
axes.plot(make_ids, max_data, c=colors[4], linewidth=4, alpha=0.7)
axes.set_xticks(range(-1, 13))
axes.set_xticklabels([" "] + geradados.get_make_labels(pddata))
axes.set_xlabel("Make", fontsize=16)
axes.set_ylabel("Price", fontsize=16)
high_patch = mpl.patches.Patch(color=colors[4], alpha=0.7, label="High")
mean_patch = mpl.patches.Patch(color=colors[3], alpha=0.7, label="Mean")
low_patch = mpl.patches.Patch(color=colors[2], alpha=0.7, label="Low")
axes.legend(handles=[high_patch, mean_patch, low_patch], loc=2)
if gs:
gs.tight_layout(figure)
return axes