def doubles_classification_stats(positions,
energies,
classification,
close_max=1.0,
scale=False):
"""Outputs calculated separation distances, relative energies,
and energy differences for double events in the dataset.
:param positions: event positions of interest, e.g validation positions
:param energies: event energies of interest, e.g validation energies
:param classification: event classification result
:param close_max: Upper limit to what is to be considered a 'close'
event. Defaults to 1.0 pixels.
:return df_doubles: DataFrame containing information about double events
"""
s_idx, d_idx, c_idx = event_indices(positions, close_max)
sep_dist = separation_distance(positions[d_idx])
energy_diff = energy_difference(energies[d_idx])
rel_energy = relative_energy(energies[d_idx], scale=scale)
df_doubles = pd.DataFrame(data={
"close": np.isin(d_idx, c_idx),
"separation distance": sep_dist.flatten(),
"relative energy": rel_energy.flatten(),
"energy difference": energy_diff.flatten(),
"classification": classification[d_idx].flatten(),
"indices": d_idx.flatten(),
},
index=np.arange(d_idx.shape[0]))
return df_doubles
def singles_classification_stats(positions, energies, classification):
"""Outputs calculated separation distances, relative energies,
and energy differences for double events in the dataset.
:param positions: event positions of interest, e.g validation positions
:param energies: event energies of interest, e.g validation energies
:param classification: event classification result
:return df_singles: DataFrame containing information about single events
"""
s_idx, d_idx, c_idx = event_indices(positions)
df_singles = pd.DataFrame(data={
"x_pos": positions[s_idx, 0].flatten(),
"y_pos": positions[s_idx, 1].flatten(),
"energy": energies[s_idx, 0].flatten(),
"classification": classification[s_idx].flatten(),
"indices": s_idx.flatten(),
},
index=np.arange(s_idx.shape[0]))
return df_singles
warnings.filterwarnings('ignore', category=FutureWarning)
# ================== Config =======================
with open("results_experiment_config.json", 'r') as fp:
config = json.load(fp)
# ================== Callbacks ====================
# ================== Import Data ==================
DATA_PATH = get_git_root() + "data/simulated/"
images = np.load(DATA_PATH + config['data']['images'])
images = images.reshape(images.shape[0], 256)
energies = np.load(DATA_PATH + config['data']['energies'])
positions = np.load(DATA_PATH + config['data']['positions'])
single_indices, double_indices, close_indices = event_indices(positions)
# log-scale the images if desireable
config['scaling'] = "minmax"
if "np.log" in config['scaling']:
images = np.log1p(images)
# set tf random seed
tf.random.set_seed(config['random_seed'])
# ================== Import Data ==================
with tf.device(get_tf_device(20)):
model = Sequential()
model.add(Dense(2, activation='relu', input_shape=(256, )))
model.compile(
loss='mse',
optimizer='adam',
)