def set_title_for(ax, im):
im = im.reshape([1, 56, 64])
energy, max_energy = get_energies(im), get_max_energy(im)
ax.set_title("Energy: %.2f GeV\nMaximum Energy: %.2f GeV" % (energy, max_energy))
folder=None,
ax=axes[func_idx],
labels=["Geant4", "Im2Im"],
**params)
build_histogram_HTOS(true=mc_data_images_m,
fake=generated_images,
energy_scaler=calo_scaler,
threshold=3600,
real_ET=tracker_real_ET,
labels=["Geant4", "Im2Im"],
ax1=axes[-3],
ax2=axes[-2])
axes[-1].scatter(tracker_real_ET,
get_energies(mc_data_images_m),
label="Geant4",
alpha=0.05)
axes[-1].scatter(tracker_real_ET,
get_energies(generated_images),
label="Im2Im",
alpha=0.05)
axes[-1].legend()
figs.append(fig2)
use_functions[get_center_of_mass_r] = {"image_shape": image_shape}
for idx in range(10):
print("Single images:", idx + 1, "/", 10)
use_functions[get_energy_resolution] = {
"real_ET": tracker_real_ET[idx],
fake=generated_images,
function=func,
name=colnames[func_idx],
epoch="",
folder=None,
ax=axes[func_idx],
**params)
build_histogram_HTOS(true=calo_images,
fake=generated_images,
energy_scaler=1,
threshold=3.6,
real_ET=tracker_et,
ax1=axes[7],
ax2=axes[8])
axes[-1].scatter(tracker_et,
get_energies(calo_images),
label="true",
alpha=0.1)
axes[-1].scatter(tracker_et,
get_energies(generated_images),
label="fake",
alpha=0.1)
axes[-1].legend()
axes[-1].set_xlim(0, np.max(tracker_et))
axes[-1].set_ylim(0, np.max(tracker_et))
axes[-1].set_xlabel("Tracker")
axes[-1].set_ylabel("Calorimeter")
figs.append(fig2)
for idx in range(nr_samples):
print(idx + 1, "/", nr_samples)
get_std_energy: {
"energy_scaler": calo_scaler / 1000,
"threshold": 5 / calo_scaler
},
get_energy_resolution: {
"real_ET": tracker_real_ET,
"energy_scaler": calo_scaler
}
}
gan_data_m = gan_data_m.reshape([-1, *image_shape[:-1]])
assert mc_data_images_m.shape == generated_images.shape == gan_data_m.shape, "Shape mismatch."
axes[model_idx,
-1].scatter(tracker_real_ET / 1000,
get_energies(mc_data_images_m,
energy_scaler=calo_scaler / 1000),
label="Geant4",
alpha=0.05)
axes[model_idx,
-1].scatter(tracker_real_ET / 1000,
get_energies(generated_images,
energy_scaler=calo_scaler / 1000),
label="Im2Im",
alpha=0.05)
axes[model_idx,
-1].scatter(tracker_real_ET / 1000,
get_energies(gan_data_m,
energy_scaler=calo_scaler / 1000),
label="CGAN",
alpha=0.05)
axes[model_idx, -1].set_xlabel("Tracker [GeV]")
get_std_energy: {
"energy_scaler": calo_scaler / 1000,
"threshold": 5 / calo_scaler
},
get_energy_resolution: {
"real_ET": tracker_real_ET,
"energy_scaler": calo_scaler
}
}
assert (mc_data_images_m.shape == generated_images_from_cgan.shape ==
generated_images_from_tracker.shape, "Shape mismatch.")
axes[model_idx,
-1].scatter(tracker_real_ET / 1000,
get_energies(mc_data_images_m,
energy_scaler=calo_scaler / 1000),
label="Geant4",
alpha=0.05)
axes[model_idx,
-1].scatter(tracker_real_ET / 1000,
get_energies(generated_images_from_cgan,
energy_scaler=calo_scaler / 1000),
label="CGAN",
alpha=0.05)
axes[model_idx,
-1].scatter(tracker_real_ET / 1000,
get_energies(generated_images_from_tracker,
energy_scaler=calo_scaler / 1000),
label="Tracker",
alpha=0.05)
axes[model_idx, -1].set_xlabel("Tracker [GeV]")
def set_title_for(ax, im, name):
im = im.reshape([1, 64, 64])
energy, max_energy = get_energies(im), get_max_energy(im)
ax.set_title("%s\nEnergy: %.2f GeV\nMaximum Energy: %.2f GeV" %
(name, energy, max_energy))
folder=None,
ax=axes[func_idx],
use_legend=True,
**params)
figs.append(fig_caloStat)
fig_calo_distribution, ax = plt.subplots(nrows=1,
ncols=3,
figsize=(12, 8),
facecolor='w',
edgecolor='k')
fig_calo_distribution.subplots_adjust(wspace=0.2, hspace=0.05)
## Distribution total energy
calo_energies = functionsOnImages.get_energies(calo_images)
ax[0].hist(calo_energies, bins=20)
ax[0].set_xlabel("Calorimeter energy")
ax[0].set_title("Calorimeter energy density")
## Distribution resolution
energy_resolution = (tracker_events["real_ET"].values -
calo_energies) / tracker_events["real_ET"].values
ax[1].hist(energy_resolution, bins=20)
ax[1].set_xlabel("(true - reco) / true")
ax[1].set_title("Calorimeter resolution")
## ECDF
unique_calo_energies = np.sort(np.unique(calo_energies))
cumulative_sum = [0] * len(unique_calo_energies)
phi1 = get_phi(x=tracker["x_projections"], y=tracker["y_projections"])
phi2 = get_phi(x=tracker["momentum_px"], y=tracker["momentum_py"])
fig_phi, _ = check_angles(phi1, phi2, "Phi")
figs.append(fig_phi)
z = 13330 + 1220 / 2
theta1 = np.arcsin(
np.sqrt(tracker["x_projections"]**2 + tracker["y_projections"]**2) /
np.sqrt(tracker["x_projections"]**2 + tracker["y_projections"]**2 +
z**2))
theta2 = np.arcsin(tracker["momentum_pt"] / tracker["momentum_p"])
fig_theta, _ = check_angles(theta1, theta2, "Theta")
figs.append(fig_theta)
tracker["CaloEnergy"] = get_energies(calo)
tracker["CaloMaxEnergy"] = get_max_energy(calo)
tracker["Cells"] = get_number_of_activated_cells(calo)
tracker["theta"] = theta1
tracker["phi"] = phi1 + np.pi
fig_scan, axs = plt.subplots(2, 4, figsize=(16, 9))
phi_range = np.linspace(0, 2 * np.pi, 5)
for i in range(4):
plot_in_range(tracker,
calo,
"phi", [phi_range[i], phi_range[i + 1]],
ax=axs[0, i])
theta_range = np.linspace(min(tracker["theta"]), max(tracker["theta"]), 5)
for i in range(4):