def plot_arcma():
arcma = ARCMA_Model()
p = 15
arcma_threshold_classification = 0.65
relevant_features = s.load_var(
"arcma_relevant_features_best_window{}relevant_features_{}.pkl".format(
slash, p))
y = s.load_var("arcma_relevant_features_best_window{}y_{}.pkl".format(
slash, p))
y = pd.DataFrame(y, columns=[arcma.label_tag])
#Remove Activit with label 0, not discribed in arcma doc
#invalid_indexes = list(y[y["activity"]==0].index)
#y_new = y.drop(labels=invalid_indexes)
#relevant_features_new = relevant_features.drop(labels=invalid_indexes)
#Translate Arcma Labels
y[y["activity"] == 0] = "Indefinida"
y[y["activity"] == 1] = "working_at_computer"
y[y["activity"] == 2] = "standing_up_walking_going_updown_stairs"
y[y["activity"] == 3] = "standing"
y[y["activity"] == 4] = "walking"
y[y["activity"] == 5] = "going_updown_stairs"
y[y["activity"] == 6] = "walking_and_talking_with_someone."
y[y["activity"] == 7] = "talking_while_standing"
balanced_data = balance_data.balance_data(relevant_features, y,
threshold_balance_data)
plot_confusion_matrix(arcma, balanced_data[0], balanced_data[1],
arcma_threshold_classification)
# -*- coding: utf-8 -*-
# IMPORTS #
from utils.debug import Debug
from models.arcma_model import ARCMA_Model
from tsfresh import extract_relevant_features
from pre_processing.processing_db_files import Processing_DB_Files
from utils.project import Project, slash
from scripts.save_workspace import save
#===INITIALIZATION===#
Debug.DEBUG = 0
arcma = ARCMA_Model()
processing = Processing_DB_Files()
project = Project()
s = save()
#window = 26 # Janela Fixa
window = 50 # Melhor Janela
persons = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]
for p in persons:
data = arcma.load_training_data_by_people(p)
print("Slicing Window....")
data_tsfresh, y = arcma.slice_by_window_tsfresh(data, window)
y.index += 1
del data_tsfresh["activity"]
classes_counts = y.value_counts()
if len(classes_counts) > 1:
relevant_features = extract_relevant_features(data_tsfresh,
y,
from sklearn.ensemble import ExtraTreesClassifier # Extra Trees
from sklearn.naive_bayes import GaussianNB #Naive Bayes
from sklearn import svm #SVM
from sklearn.neural_network import MLPClassifier #multi-layer percept
import pandas as pd
from sklearn.model_selection import train_test_split
from pre_processing.get_accuracy import Get_Accuracy
from scripts.save_workspace import save
import numpy as np
from pre_processing.balance_data import BalanceData
import statistics as st
#===INITIALIZATION===#
Debug.DEBUG = 0
arcma = ARCMA_Model()
processing = Processing_DB_Files()
project = Project()
#tuple from MPL
t_aux = []
for i in range(0,500):
t_aux.append(500)
t = tuple(t_aux)
####
classifiers = {"MPL": MLPClassifier(random_state=1, solver="adam", activation="relu", max_iter=100000, alpha=1e-5, hidden_layer_sizes=t), "Extratrees": ExtraTreesClassifier(n_estimators = 1000, random_state=1), "Knn":KNeighborsClassifier(n_neighbors=5), "Naive Bayes":GaussianNB(), "RandomForest":RandomForestClassifier(n_estimators = 1000, random_state=1), "Decision Tree":tree.DecisionTreeClassifier(random_state=1), "SVM":svm.SVC(probability=True, random_state=1)}
persons = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
get_accuracy = Get_Accuracy()
balance_data = BalanceData()
threshold_balance_data = 40
#Select the best classifier
accuracy_mean = pd.DataFrame(columns=["Classifier", "Accuracy"])
#Round 01
models.append({
"model_name": "hmp",
"model": HMP_Model(),
"persons": hmp_persons,
"window": 16
})
models.append({
"model_name": "umafall",
"model": UMAFALL_Model(),
"persons": umafall_persons,
"window": 10
})
models.append({
"model_name": "arcma",
"model": ARCMA_Model(),
"persons": arcma_persons,
"window": 26
})
#Round 02
#models.append({"model_name":"hmp", "model":HMP_Model(), "persons":hmp_persons, "window":90})
#models.append({"model_name":"umafall", "model":UMAFALL_Model(), "persons":umafall_persons, "window":10})
#models.append({"model_name":"arcma", "model":ARCMA_Model(), "persons":arcma_persons, "window":40})
#tuple from MPL
t_aux = []
for i in range(0, 500):
t_aux.append(500)
t = tuple(t_aux)
####
classifiers = {
# -*- coding: utf-8 -*-
from models.arcma_model import ARCMA_Model
from models.hmp_model import HMP_Model
from models.umafall_model import UMAFALL_Model
#ARCMA
arcma = ARCMA_Model()
arcma.load_training_data_by_window_by_people(1, 36)
sample_arcma = arcma.training[0]
sample_arcma_to_plot = sample_arcma.drop("activity", 1)
sample_arcma_to_plot.plot(subplots=True, legend=True, figsize=(20, 10))
#HMP
hmp = HMP_Model()
hmp.load_training_data_by_window_by_people("f1", 36)
sample_hmp = hmp.training[0]
sample_hmp_to_plot = sample_hmp.drop("activity", 1)
sample_hmp_to_plot.plot(subplots=True, legend=True, figsize=(20, 10))
#UMAFALL
umafall = UMAFALL_Model()
umafall.load_training_data_by_window_by_people(
"1", 10, additional_where="and sensor=2 and body=3")
sample_umafall = umafall.training[0]
sample_umafall_to_plot = sample_umafall.drop("activity", 1)
sample_umafall_to_plot.plot(subplots=True, legend=True, figsize=(20, 10))
#===INIT BASES===#
hmp_persons = ["f1", "m1", "m2", "f2", "m3", "f3", "m4", "f4"] # at least 5 activities
#umafall_persons = [1,2,3,4,5,6,8,9,10,11,12,13,14,15,16,17]
umafall_persons = [8,10,11,12,13,14,15,16,17]
arcma_persons = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
models = []
#Round 01
#models.append({"model_name":"hmp", "model":HMP_Model(), "persons":hmp_persons, "window":16, "where":""})
#models.append({"model_name":"umafall", "model":UMAFALL_Model(), "persons":umafall_persons, "window":10, "where": "and body=2 and sensor=2"}) #MAGNETOMETER in WAIST
#models.append({"model_name":"arcma", "model":ARCMA_Model(), "persons":arcma_persons, "window":26, "where":""})
#Round 02
models.append({"model_name":"hmp", "model":HMP_Model(), "persons":hmp_persons, "window":90, "where":""})
#models.append({"model_name":"umafall", "model":UMAFALL_Model(), "persons":umafall_persons, "window":10, "where": "and body=2 and sensor=2"}) #MAGNETOMETER in WAIST
models.append({"model_name":"arcma", "model":ARCMA_Model(), "persons":arcma_persons, "window":40, "where":""})
for model in models:
for p in model["persons"]:
try:
data = model['model'].load_training_data_by_people(p, additional_where = model['where'])
print("Slicing Window....")
data_tsfresh, y = model['model'].slice_by_window_tsfresh(data, model["window"])
y.index += 1
del data_tsfresh["activity"]
classes_counts = y.value_counts()
if len(classes_counts) > 1:
relevant_features = extract_relevant_features(data_tsfresh, y, column_id='id', column_sort='time')
s.save_var(relevant_features, "new_features{}{}_relevant_features_window_{}{}relevant_features_{}.pkl".format(slash, model['model_name'], model['window'], slash, p))
s.save_var(y, "new_features{}{}_relevant_features_window_{}{}y_{}.pkl".format(slash, model['model_name'], model['window'], slash, p))