# Author : Amir Shokri
# github link : https://github.com/amirshnll/Vicon-Physical-Action
# dataset link : http://archive.ics.uci.edu/ml/datasets/Vicon+Physical+Action+Data+Set
# email : amirsh.nll@gmail.com

Read data

import pandas as pd
import numpy as np
import os

path_data='data'
lbl_binary=list()
lbl_20_class=list()
data=[]
for folder in os.listdir(path_data):
    path_in=os.path.join(path_data,folder)
    for sub_folder in os.listdir(path_in):
        
        print('---------{0}------------'.format(sub_folder))
        path_in2=os.path.join(path_in,sub_folder)
        for file_name in os.listdir(path_in2):
            lbl_one=file_name[:-4]
            print(file_name[:-4])
            path_file=os.path.join(path_in2,file_name)
            txt_data=np.array(pd.read_csv(path_file, delim_whitespace=True))
            txt_data=txt_data[:,1:]
            data.extend(txt_data)
            lbl_all=((lbl_one+',')*txt_data.shape[0]).split(',')[:-1]
            lbl_20_class.extend(lbl_all)
            print(txt_data.shape)

---------aggressive------------
Elbowing
(2396, 27)
Frontkicking
(2091, 27)
Hamering
(1901, 27)
Headering
(1510, 27)
Kneeing
(2460, 27)
Pulling
(1872, 27)
Punching
(2485, 27)
Pushing
(2095, 27)
Sidekicking
(1444, 27)
Slapping
(2136, 27)
---------normal------------
Bowing
(2013, 27)
Clapping
(1715, 27)
Handshaking
(2558, 27)
Hugging
(1972, 27)
Jumping
(1906, 27)
Running
(2241, 27)
Seating
(2273, 27)
Standing
(2101, 27)
Walking
(1720, 27)
Waving
(971, 27)
---------aggressive------------
Elbowing
(2093, 27)
Frontkicking
(2133, 27)
Hamering
(1933, 27)
Headering
(1956, 27)
Kneeing
(2471, 27)
Pulling
(1788, 27)
Punching
(1858, 27)
Pushing
(1639, 27)
Sidekicking
(2425, 27)
Slapping
(2462, 27)
---------normal------------
Bowing
(1943, 27)
Clapping
(1822, 27)
Handshaking
(1591, 27)
Hugging
(2019, 27)
Jumping
(2030, 27)
Running  [worse]
(2093, 27)
Running
(2338, 27)
Seating
(2096, 27)
Standing
(2463, 27)
Walking
(1555, 27)
---------aggressive------------
Elbowing
(2012, 27)
Frontkicking
(1675, 27)
Hamering
(1465, 27)
Headering
(2473, 27)
Kneeing
(1892, 27)
Pulling
(2348, 27)
Punching
(2099, 27)
Pushing
(1483, 27)
Sidekicking
(2271, 27)
Slapping
(1714, 27)
---------normal------------
Bowing
(1842, 27)
Clapping
(1983, 27)
Handshaking
(1829, 27)
Hugging
(2010, 27)
Jumping
(1099, 27)
Running
(2287, 27)
Seating
(2325, 27)
Standing
(2352, 27)
Walking
(1685, 27)
Waving
(1963, 27)
---------aggressive------------
Elbowing
(3532, 27)
Frontkicking
(3314, 27)
Hamering
(3926, 27)
Headering
(3832, 27)
Kneeing
(3513, 27)
Pulling
(3622, 27)
Punching
(3557, 27)
Pushing
(3126, 27)
Sidekicking
(2335, 27)
Slapping
(3522, 27)
---------normal------------
Bowing
(3673, 27)
Clapping
(3609, 27)
Handshaking
(3029, 27)
Hugging
(3307, 27)
Jumping
(2957, 27)
Running
(1534, 27)
Seating
(3021, 27)
Standing
(3162, 27)
Walking
(2005, 27)
Waving
(2769, 27)
---------aggressive------------
Elbowing
(2804, 27)
Frontkicking
(3324, 27)
Hamering
(3574, 27)
Headering
(4013, 27)
Kneeing
(2977, 27)
Pulling
(4235, 27)
Punching
(3419, 27)
Pushing
(3413, 27)
Sidekicking
(4193, 27)
Slapping
(4223, 27)
---------normal------------
Bowing
(3751, 27)
Clapping
(3045, 27)
Handshaking
(3078, 27)
Hugging
(4103, 27)
Jumping
(2946, 27)
Running
(3032, 27)
Seating
(3763, 27)
Standing
(4106, 27)
Walking
(3087, 27)
Waving
(6416, 27)
---------aggressive------------
Elbowing
(4119, 27)
Frontkicking
(4146, 27)
Hamering
(6242, 27)
Headering
(3706, 27)
Kneeing
(5516, 27)
Pulling
(6399, 27)
Punching
(4232, 27)
Pushing
(4260, 27)
Sidekicking
(4174, 27)
Slapping
(6519, 27)
---------normal------------
Bowing
(4272, 27)
Clapping
(3952, 27)
Handshaking
(4197, 27)
Hugging
(4335, 27)
Jumping
(4054, 27)
Running
(4269, 27)
Seating
(4244, 27)
Standing
(3838, 27)
Walking
(6373, 27)
Waving
(4305, 27)
---------aggressive------------
Elbowing
(3649, 27)
Frontkicking
(4365, 27)
Hamering
(4138, 27)
Headering
(3695, 27)
Kneeing
(5972, 27)
Pulling
(3595, 27)
Punching
(4757, 27)
Pushing
(2831, 27)
Sidekicking
(3689, 27)
Slapping
(6255, 27)
---------normal------------
Bowing
(4155, 27)
Clapping
(4733, 27)
Handshaking
(4179, 27)
Hugging
(4763, 27)
Jumping
(4142, 27)
Running
(4154, 27)
Seating
(6273, 27)
Standing
(4215, 27)
Walking
(5949, 27)
Waving
(4035, 27)
---------aggressive------------
Elbowing
(1691, 27)
Frontkicking
(1925, 27)
Hamering
(1893, 27)
Headering
(1885, 27)
Kneeing
(1793, 27)
Pulling
(1988, 27)
Punching
(2167, 27)
Pushing
(1508, 27)
Sidekicking
(1986, 27)
Slapping
(1954, 27)
---------normal------------
Bowing
(1944, 27)
Clapping
(2066, 27)
Handshaking
(1778, 27)
Hugging
(1624, 27)
Jumping
(2285, 27)
Running
(1059, 27)
Seating
(1888, 27)
Standing
(1999, 27)
Walking
(1645, 27)
Waving
(2193, 27)
---------aggressive------------
Elbowing
(4477, 27)
Frontkicking
(4122, 27)
Hamering
(6362, 27)
Headering
(6541, 27)
Kneeing
(3182, 27)
Pulling
(4484, 27)
Punching
(4380, 27)
Pushing
(4288, 27)
Sidekicking
(4201, 27)
Slapping
(4293, 27)
---------normal------------
Bowing
(4505, 27)
Clapping
(4441, 27)
Handshaking
(4433, 27)
Hugging
(6607, 27)
Jumping
(4215, 27)
Running
(4299, 27)
Seating
(6779, 27)
Standing
(3806, 27)
Walking
(4426, 27)
Waving
(4048, 27)
---------aggressive------------
Elbowing
(1510, 27)
Frontkicking
(1474, 27)
Hamering
(782, 27)
Headering
(2115, 27)
Kneeing
(1622, 27)
Pulling
(1142, 27)
Punching
(1807, 27)
Pushing
(1681, 27)
Sidekicking
(792, 27)
Slapping
(2155, 27)
---------normal------------
Bowing
(1590, 27)
Clapping
(1928, 27)
Handshaking
(2000, 27)
Hugging
(2189, 27)
Jumping
(1671, 27)
Running
(1632, 27)
Seating
(2337, 27)
Standing
(2176, 27)
Walking
(2302, 27)
Waving
(2198, 27)

print(len(lbl_20_class))
print(len(data))

605181
605181

Preproccessing

print('-----------Preproccessing----------------')

from sklearn.preprocessing import LabelEncoder
import numpy as np
encoder = LabelEncoder()
lbl_20_class=np.array(lbl_20_class)
lbl_20_class=encoder.fit_transform(lbl_20_class.reshape(-1, 1))



print(lbl_20_class)

-----------Preproccessing----------------


C:\Users\Human\anaconda3\lib\site-packages\sklearn\utils\validation.py:73: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().
  return f(**kwargs)


[ 2  2  2 ... 20 20 20]

normalize

from sklearn import preprocessing
Data_lable=lbl_20_class
Data_main=data
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(Data_main)
Data_main=scaler.transform(Data_main)
print(Data_main)

[[0.68426919 0.55894291 0.70059154 ... 0.6955238  0.51383589 0.10203499]
 [0.68426919 0.55894291 0.70059154 ... 0.6955238  0.51383589 0.10203499]
 [0.68426919 0.55894291 0.70059154 ... 0.6955238  0.51383589 0.10203499]
 ...
 [0.4956291  0.48598417 0.73733102 ... 0.52708659 0.57787027 0.09402593]
 [0.4956291  0.48598417 0.73733102 ... 0.52708659 0.57787027 0.09402593]
 [0.4956291  0.48598417 0.73733102 ... 0.52708659 0.57787027 0.09402593]]

'DT','KNN','NB','MLP','LR'

import numpy as np
import os
from pytictoc import TicToc
import matplotlib.pyplot as plt
from sklearn.metrics import classification_report, confusion_matrix
from sklearn.metrics import accuracy_score
from sklearn.neighbors import KNeighborsClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.naive_bayes import BernoulliNB,MultinomialNB
from sklearn.linear_model import LogisticRegression
from sklearn.neural_network import MLPClassifier


def run_all_algorithm(Train_data,Test_data,Train_lable,Test_lable,str_out):
    
        
    print(np.shape(Test_data))
    print(np.shape(Train_data))
    print(np.shape(Train_lable))
    print(np.shape(Test_lable))
    
    algorithms_name=['DT','KNN','NB','MLP','LR']
    alg_num=len(algorithms_name)
    accuracy_array=np.zeros(alg_num)
    precision_array=np.zeros(alg_num)
    recall_array=np.zeros(alg_num)
    f1_score_array=np.zeros(alg_num)
    time_array=np.zeros(alg_num)

    t = TicToc()
          

    print('---------------------DT---------------------')
    K=0;
    t.tic() #Start timer
    classifier_DT = DecisionTreeClassifier(max_depth=1000,random_state=0)
    classifier_DT.fit(Train_data, Train_lable)
    Test_predict = classifier_DT.predict(Test_data)
    Con_matrix=confusion_matrix(Test_lable, Test_predict)
    TimeDT=t.tocvalue() #Time elapsed since t.tic()
    classfi_report=classification_report(Test_lable, Test_predict,output_dict=True)
    
    # save to array
    accuracy_array[K]=accuracy_score(Test_lable, Test_predict)
    precision_array[K]= classfi_report['macro avg']['precision'] 
    recall_array[K]= classfi_report['macro avg']['recall']    
    f1_score_array[K]= classfi_report['macro avg']['f1-score']
    time_array[K]=TimeDT
    print(accuracy_array[K])
    
    
    print('--------------NB----------------')
    K+=1;
    t.tic() #Start timer
    classifier = MultinomialNB()
    classifier.fit(Train_data, Train_lable)
    Test_predict = classifier.predict(Test_data)
    TimeNB=t.tocvalue() #Time elapsed since t.tic()
    Con_matrix=confusion_matrix(Test_lable, Test_predict)
    classfi_report=classification_report(Test_lable, Test_predict,output_dict=True)
    
    
    # save to array
    accuracy_array[K]=accuracy_score(Test_lable, Test_predict)
    precision_array[K]= classfi_report['macro avg']['precision'] 
    recall_array[K]= classfi_report['macro avg']['recall']    
    f1_score_array[K]= classfi_report['macro avg']['f1-score']
    time_array[K]=TimeNB
    print(accuracy_array[K])
    print('---------------------KNN---------------------')
    K+=1;
    t.tic() #Start timer
    classifier=KNeighborsClassifier(n_neighbors=10)
    classifier.fit(Train_data, Train_lable)
    Test_predict = classifier.predict(Test_data)
    TimeKNN=t.tocvalue() #Time elapsed since t.tic()
    Con_matrix=confusion_matrix(Test_lable, Test_predict)
    classfi_report=classification_report(Test_lable, Test_predict,output_dict=True)
    
    # save to array
    accuracy_array[K]=accuracy_score(Test_lable, Test_predict)
    precision_array[K]= classfi_report['macro avg']['precision'] 
    recall_array[K]= classfi_report['macro avg']['recall']    
    f1_score_array[K]= classfi_report['macro avg']['f1-score']
    time_array[K]=TimeKNN
    print(accuracy_array[K])
    
    print('---------------------MLP---------------------')
    K+=1;
    t.tic() #Start timer
    classifier=MLPClassifier( solver='adam', random_state=0,hidden_layer_sizes=[5], max_iter=200)
    classifier.fit(Train_data, Train_lable)
    Test_predict = classifier.predict(Test_data)
    TimeMLP=t.tocvalue() #Time elapsed since t.tic()
    Con_matrix=confusion_matrix(Test_lable, Test_predict)
    classfi_report=classification_report(Test_lable, Test_predict,output_dict=True)
    
    # save to array
    accuracy_array[K]=accuracy_score(Test_lable, Test_predict)
    precision_array[K]= classfi_report['macro avg']['precision'] 
    recall_array[K]= classfi_report['macro avg']['recall']    
    f1_score_array[K]= classfi_report['macro avg']['f1-score']
    time_array[K]=TimeMLP
    print(accuracy_array[K])
    
    print('---------------------LogisticRegression---------------------')
    K+=1;
    t.tic() #Start timer
    classifier=LogisticRegression()
    classifier.fit(Train_data, Train_lable)
    Test_predict = classifier.predict(Test_data)
    TimeLR=t.tocvalue() #Time elapsed since t.tic()
    Con_matrix=confusion_matrix(Test_lable, Test_predict)
    classfi_report=classification_report(Test_lable, Test_predict,output_dict=True)
    
    # save to array
    accuracy_array[K]=accuracy_score(Test_lable, Test_predict)
    precision_array[K]= classfi_report['macro avg']['precision'] 
    recall_array[K]= classfi_report['macro avg']['recall']    
    f1_score_array[K]= classfi_report['macro avg']['f1-score']
    time_array[K]=TimeLR
    print(accuracy_array[K])
    
    
    
    
    

    H=6
    L=8
    
    
    
    print('--------------------result--------------------------')
    fig1=plt.figure(figsize=(H, L)) #  
    plt.bar(algorithms_name, accuracy_array,color = ['red', 'green'])
    plt.xticks(algorithms_name, rotation=70)
    plt.ylabel('percent%')
    plt.title('Accuracy of all Algorithm')
    plt.xlabel("Algoritm names")
    for i, v in enumerate(accuracy_array):
        v=round(v,2)
        plt.text(i-0.2 , v+0.01 , str(v), color='blue', fontweight='bold')
    fig1.show()
    plt.savefig(os.path.join(str_out+' accuracy.png'), dpi=300, format='png', bbox_inches='tight') # use format='svg' or 'pdf' for vectorial pictures
    
     
    fig2=plt.figure(figsize=(H, L)) #  
    plt.bar(algorithms_name, precision_array,color = ['red', 'green'])
    plt.xticks(algorithms_name, rotation=70)
    plt.ylabel('percent%')
    plt.title('Precision of all Algorithm')
    plt.xlabel("Algoritm names")
    for i, v in enumerate(precision_array):
        v=round(v,2)
        plt.text(i-0.2 , v+0.01 , str(v), color='blue', fontweight='bold')
    fig2.show()
    plt.savefig(os.path.join(str_out+' precision.png'), dpi=300, format='png', bbox_inches='tight') # use format='svg' or 'pdf' for vectorial pictures
    
    
    
    
    fig3=plt.figure(figsize=(H, L)) #  
    plt.bar(algorithms_name, recall_array,color = ['red', 'green'])
    plt.xticks(algorithms_name, rotation=70)
    plt.ylabel('percent%')
    plt.title('Recallof all Algorithm')
    plt.xlabel("Algoritm names")
    for i, v in enumerate(recall_array):
        v=round(v,2)
        plt.text(i-0.2 , v+0.01 , str(v), color='blue', fontweight='bold')
    fig3.show()
    plt.savefig(os.path.join(str_out+' recall.png'), dpi=300, format='png', bbox_inches='tight') # use format='svg' or 'pdf' for vectorial pictures
    
    
    
    fig4=plt.figure(figsize=(H, L)) #  
    plt.bar(algorithms_name, f1_score_array,color = ['red', 'green'])
    plt.xticks(algorithms_name, rotation=70)
    plt.ylabel('percent%')
    plt.title('f1-score of all Algorithm')
    plt.xlabel("Algoritm names")
    for i, v in enumerate(f1_score_array):
        v=round(v,2)
        plt.text(i-0.2 , v+0.01 , str(v), color='blue', fontweight='bold')
    fig4.show()
    plt.savefig(os.path.join(str_out+' f1_score.png'), dpi=300, format='png', bbox_inches='tight') # use format='svg' or 'pdf' for vectorial pictures
    
    
    
    fig5=plt.figure(figsize=(H, L)) # 
    plt.bar(algorithms_name, time_array,color = ['blue', 'green'])
    plt.xticks(algorithms_name, rotation=70)
    plt.ylabel('time(s)')
    plt.title('time of all Algorithm')
    plt.xlabel("Algoritm names")
    for i, v in enumerate(time_array):
        v=round(v,2)
        plt.text(i-0.2 , v+0.01 , str(v), color='blue', fontweight='bold')
    plt.savefig(os.path.join(str_out+' time.png'), dpi=300, format='png', bbox_inches='tight') # use format='svg' or 'pdf' for vectorial pictures
    fig5.show()
    
    
    np.savetxt(str_out+'accuracy.csv', accuracy_array, delimiter=',')
    np.savetxt(str_out+' precision_array.csv', precision_array, delimiter=',')
    np.savetxt(str_out+'recall_array.csv', recall_array, delimiter=',')
    np.savetxt(str_out+' time_array.csv', time_array, delimiter=',')
    np.savetxt(str_out+' f1-score.csv', f1_score_array, delimiter=',')

train_test_split

from sklearn.model_selection import train_test_split
Train_data, Test_data, Train_lable, Test_lable = train_test_split(Data_main, Data_lable, test_size=0.20)
run_all_algorithm(Train_data, Test_data, Train_lable, Test_lable ,"")

(121037, 27)
(484144, 27)
(484144,)
(121037,)
---------------------DT---------------------
0.9986119946793129
--------------NB----------------


C:\Users\Human\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, msg_start, len(result))


0.07363037748787561
---------------------KNN---------------------
0.998562423060717
---------------------MLP---------------------


C:\Users\Human\anaconda3\lib\site-packages\sklearn\neural_network\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.
  warnings.warn(
C:\Users\Human\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.
  _warn_prf(average, modifier, msg_start, len(result))


0.48655369845584406
---------------------LogisticRegression---------------------


C:\Users\Human\anaconda3\lib\site-packages\sklearn\linear_model\_logistic.py:762: ConvergenceWarning: lbfgs failed to converge (status=1):
STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.

Increase the number of iterations (max_iter) or scale the data as shown in:
    https://scikit-learn.org/stable/modules/preprocessing.html
Please also refer to the documentation for alternative solver options:
    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression
  n_iter_i = _check_optimize_result(


0.5379594669398614
--------------------result--------------------------


<ipython-input-60-a1fc3c3f9089>:144: UserWarning: Matplotlib is currently using module://ipykernel.pylab.backend_inline, which is a non-GUI backend, so cannot show the figure.
  fig1.show()
<ipython-input-60-a1fc3c3f9089>:157: UserWarning: Matplotlib is currently using module://ipykernel.pylab.backend_inline, which is a non-GUI backend, so cannot show the figure.
  fig2.show()
<ipython-input-60-a1fc3c3f9089>:172: UserWarning: Matplotlib is currently using module://ipykernel.pylab.backend_inline, which is a non-GUI backend, so cannot show the figure.
  fig3.show()
<ipython-input-60-a1fc3c3f9089>:186: UserWarning: Matplotlib is currently using module://ipykernel.pylab.backend_inline, which is a non-GUI backend, so cannot show the figure.
  fig4.show()
<ipython-input-60-a1fc3c3f9089>:201: UserWarning: Matplotlib is currently using module://ipykernel.pylab.backend_inline, which is a non-GUI backend, so cannot show the figure.
  fig5.show()

semnan-university-ai/Vicon-Physical-Action

Read data

Preproccessing

normalize

'DT','KNN','NB','MLP','LR'

train_test_split