"""
from time import time
import dataset_io as io
import manifold
#%%
dataset = io.Dataset(augment=False)
class_to_index, sample_per_class = dataset.load_data(
path="../FOB valid 298x224", shape=(99, 66))
imgs_origin = dataset.imgs_origin
labels_origin = dataset.labels_origin
labels_origin = io.label_str2index(labels_origin, class_to_index)
imgs = imgs_origin
labels = labels_origin
#%%
import manifold
manifold_args = dict(
# Random=True,
# RandomTrees=True,
# MDS=True,
# PCA=True,
# LinearDiscriminant=True,
# Isomap=False,
# Spectral=True,
# LLE=False,
import dataset_io as io
import metrics
# %%
dataset = io.Dataset(augment=False)
class_to_index, sample_per_class = dataset.load_data(
path="dataset_1_1_origin.h5", shape=(1, 1))
imgs_train, labels_train, imgs_valid, labels_valid = dataset.train_test_split(
test_shape=0.2)
imgs_train, labels_train, imgs_valid, labels_valid, names_valid = dataset.cross_split(
total_splits=3, valid_split=0)
labels_train = io.label_str2index(labels_train, class_to_index)
labels_valid = io.label_str2index(labels_valid, class_to_index)
# Class index: {"C1": 0, "C2": 1, "C3": 2, "C4": 3, "C5": 4, "N1": 5, "N2": 6, "N3": 7, "N4": 8, "N5": 9, "N6": 10, "P1": 11, "P2": 12, "P3": 13, "P4": 14, "P5": 15}
# Sample per class: {"C5": 95, "N4": 181, "P2": 2, "C1": 220, "N2": 83, "P5": 20, "N1": 24, "C4": 66, "P4": 18, "N6": 89, "P1": 6, "C2": 60, "N3": 281, "P3": 6, "C3": 120, "N5": 83}
# %%
f = open("20180925_result.h5", "rb")
contact = load(f)
#EPOCHS=contact["epochs"],
history = contact["history"]
labels_valid = contact["labels_valid"]
mean = contact["mean"]
#names_valid2 = contact["names_valid"]
scores_predict = contact["scores_predict"]