def run(filename):
w_a1 = 0.90291
w_xy_rt = 0.010809
w_z1 = 0.357996
w_z2 = 0.229602
c_rt1 = 1.330075
c_rt2 = 1.92522
model = models.UnrollingHelicesRt2(
dbscan_features=["sina1", "cosa1", "z1", "z2", "x_rt", "y_rt"],
dbscan_weight=[w_a1, w_a1, w_z1, w_z2, w_xy_rt, w_xy_rt])
model.coef_rt1 = c_rt1
model.coef_rt2 = c_rt2
model.niter = 150
path_to_input = os.path.join(path_to_trackml, "test")
dataset_submission = []
for event_id, hits in load_dataset(path_to_input, parts=["hits"]):
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
dataset_submission.append(one_submission)
submission = pd.concat(dataset_submission)
submission.to_csv(filename, index=None)
def Fun4BO(w1, w2, w3, niter):
model.dbscan_weight[0] = w1
model.dbscan_weight[1] = w1
model.dbscan_weight[2] = w2
model.dbscan_weight[3] = w3
model.niter = int(niter)
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
score = score_event(truth, one_submission)
return score
def run(filename):
model = models.UnrollingHelices(use_outlier=False,
dbscan_features = ["sina1", "cosa1", "z1", "x1", "x2", "x_y", "x_rt", "y_rt"],
dbscan_weight = [1.0, 1.0, 0.75, 0.5, 0.5, 0.2, 0.2, 0.2])
path_to_input = os.path.join(path_to_trackml, "train_1")
for event_id, hits, truth in load_dataset(path_to_input, parts=["hits", "truth"],
skip=0, nevents=1):
def Fun4BO(w_a1, w_z1 w_x1, w_x2, w_x_y, w_xy_rt, niter):
model.dbscan_weight[0] = w_a1
model.dbscan_weight[1] = w_a1
model.dbscan_weight[2] = w_z1
model.dbscan_weight[3] = w_x1
model.dbscan_weight[4] = w_x2
model.dbscan_weight[5] = w_x_y
model.dbscan_weight[6] = w_xy_rt
model.dbscan_weight[7] = w_xy_rt
model.iter_size_helix = int(niter)
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
score = score_event(truth, one_submission)
return score
print("Bayesian Optimization")
opt = BayesianOptimization(Fun4BO,
{"w1": (0.9, 1.2),
"w2": (0.3, 0.8),
"w3": (0.1, 0.6),
"w4": (0.1, 0.6),
"w5": (0.1, 0.6),
"w6": (0.1, 0.6),
"niter": (140, 190)}, #(140, 190)
verbose = True)
opt.maximize(init_points = 3,
n_iter = 20,
acq = "ucb",
kappa = 2.576)
# [string]
labels = opt.res["max"]["max_params"].keys()
# [dict(string, [float])]
params = opt.res["all"]["params"]
len_params = len(params)
data_dic = {}
for label in labels:
val = [opt.res["max"]["max_params"][label]]
for i in range(len_params):
val.append(params[i][label])
data_dic[label] = val
data_dic["value"] = [opt.res["max"]["max_val"]] + opt.res["all"]["values"]
data_dic["label"] = ["max"] + [str(x) for x in range(len_params)]
df = pd.DataFrame(data_dic)
df.to_csv(filename, label=None)
def Fun4BO(w_a1, w_z1, w_z2, w_xy_rt, niter):
model.dbscan_weight[0] = w_a1
model.dbscan_weight[1] = w_a1
model.dbscan_weight[2] = w_z1
model.dbscan_weight[3] = w_z2
model.dbscan_weight[4] = w_xy_rt
model.dbscan_weight[5] = w_xy_rt
model.niter = int(niter)
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
score = score_event(truth, one_submission)
return score
def run(model, test_or_train, path_to_out, nevents=None):
if(test_or_train not in ["test", "train_1"]):
sys.stderr.write("Error. test_or_train must be \"test\" or \"train_1\"\n")
sys.exit()
if(test_or_train=="test" and (nevents is not None)):
sys.strerr.write("Error")
sys.exit()
path_to_input = os.path.join(path_to_trackml, test_or_train)
os.makedirs(path_to_out, exist_ok=True)
print("calculation begin : {0}".format(datetime.datetime.today()))
dataset_submission = []
if(test_or_train == "test"):
for event_id, hits in load_dataset(path_to_input, parts=["hits"]):
sys.stderr.write("processing event_id : {0}".format(event_id))
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
dataset_submission.append(one_submission)
else:
dataset_score = []
for event_id, hits, truth in load_dataset(path_to_input, parts=["hits", "truth"],
skip=0, nevents=nevents):
sys.stderr.write("processing event_id : {0}".format(event_id))
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
dataset_submission.append(one_submission)
score = score_event(truth, one_submission)
dataset_score.append(score)
print("Score for event %d:%.8f" % (event_id, score))
print("Mean Score : %.8f" % (np.sum(dataset_score)/len(dataset_score)))
submission = pd.concat(dataset_submission)
submission.to_csv(os.path.join(path_to_out, "submission.csv"), index=None)
print("calculation end : {0}".format(datetime.datetime.today()))
def Fun4BO(w_a1, w_z1, w_z2, w_xy_rt):
model.dbscan_weight[0] = w_a1
model.dbscan_weight[1] = w_a1
model.dbscan_weight[2] = w_z1
model.dbscan_weight[3] = w_z2
model.dbscan_weight[4] = w_xy_rt
model.dbscan_weight[5] = w_xy_rt
score_list = []
for (hits, truth) in zip(hits_list, truth_list):
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
score = score_event(truth, one_submission)
score_list.append(score)
return np.sum(score_list)/len(score_list)
def Fun4BO(w_a1, w_z1, w_z2, w_xy, w_xy_rt, c_r1, c_r2):
model.dbscan_weight[0] = w_a1
model.dbscan_weight[1] = w_a1
model.dbscan_weight[2] = w_z1
model.dbscan_weight[3] = w_z2
model.dbscan_weight[4] = w_xy
model.dbscan_weight[5] = w_xy_rt
model.dbscan_weight[6] = w_xy_rt
model.coef_rt1 = c_r1
model.coef_rt2 = c_r2
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
score = score_event(truth, one_submission)
return score
def Fun4BO(w_a1, w_z1, w_z2, w_xy_rt, c_rt1, c_rt2, eps0, step_eps):
model.dbscan_weight[0] = w_a1
model.dbscan_weight[1] = w_a1
model.dbscan_weight[2] = w_z1
model.dbscan_weight[3] = w_z2
model.dbscan_weight[4] = w_xy_rt
model.dbscan_weight[5] = w_xy_rt
model.coef_rt1 = c_rt1
model.coef_rt2 = c_rt2
model.eps0 = eps0
model.step_eps = step_eps
labels = model.predict(hits)
one_submission = create_one_event_submission(
event_id, hits, labels)
score = score_event(truth, one_submission)
return score
def fit(dfh, y=None):
dfh["s1"] = dfh.hit_id
dfh["N1"] = 1
dfh['r'] = np.sqrt(dfh['x'].values**2 + dfh['y'].values**2 +
dfh['z'].values**2)
dfh['rt'] = np.sqrt(dfh['x'].values**2 + dfh['y'].values**2)
dfh['a0'] = np.arctan2(dfh['y'].values, dfh['x'].values)
mm = 1
for ii in tqdm(range(self.niter), total=self.niter):
# unroll helices
mm = mm * (-1)
dfh['z1'] = (dfh['z'].values + dj) / dfh['rt'].values
dfh['z2'] = (dfh['z'].values + dj) / dfh['r'].values
dfh["a1"] = dfh.a0 + mm * (self.coef_rt * dfh.rt.values
) / 1000.0 * (ii / 2) / 180.0 * np.pi
dfh["sina1"] = np.sin(dfh["a1"].values)
dfh["cosa1"] = np.cos(dfh["a1"].values)
# scaling
ss = StandardScaler()
dfs = ss.fit_transform(dfh[self.features].values)
dfs[:, :] = dfs[:, :] * self.weight[np.newaxis, :]
# clustering
res = DBSCAN(eps=self.eps0,
min_samples=1,
metric='euclidean',
n_jobs=4).fit(dfs).labels_
dfh["s2"] = res
dfh['N2'] = dfh.groupby('s2')['s2'].transform('count')
maxs1 = np.max(dfh.s1)
dfh.s1 = np.where((dfh.N2 > dfh.N1) & (dfh.N2 < 20),
dfh.s2 + maxs1, dfh.s1)
dfh['s1'] = dfh['s1'].astype('int64')
dfh['N1'] = dfh.groupby('s1')['s1'].transform('count')
labels = dfh["s1"]
submission = create_one_event_submission(0, dfh, labels)
for i in range(self.nextend):
submission = extend(submission, dfh)
self.submission = submission
return self
if __name__ == "__main__":
sys.path.append(path_to_trackmllib)
from trackml.dataset import load_dataset
from trackml.score import score_event
os.makedirs(path_to_out, exist_ok=True)
f_log = open(path_to_log, "w")
f_log.write("calculation begin\n")
f_log.write(str(datetime.datetime.today()) + "\n")
dataset_submission = []
dataset_score = []
for event_id, hits, cells, particles, truth in load_dataset(
path_to_input, skip=0, nevents=nevents):
labels = model.predict(hits)
one_submission = create_one_event_submission(event_id, hits, labels)
dataset_submission.append(one_submission)
score = score_event(truth, one_submission)
dataset_score.append(score)
f_log.write("Score for event %d:%.8f\n" % (event_id, score))
submission = pd.concat(dataset_submission)
submission.to_csv(os.path.join(path_to_out, "submission.csv"), index=None)
f_log.write("Mean Score : %.8f\n" %
(np.sum(dataset_score) / len(dataset_score)))
f_log.write("calculation end\n")
f_log.write(str(datetime.datetime.today()))