forked from Kearlay/research
-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
7 changed files
with
724 additions
and
419 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,42 +1,30 @@ | ||
| ''' | ||
| Name: eeg_data_downloads.py | ||
| Author: Jim Chung | ||
| Dependencies: re, request, pathlib, urllib, os | ||
| Description: Executing this script will generate folders and start downloading | ||
| PhysioNet motor imagery data. Please refer to the official descriptions for | ||
| the dataset. Labels and recording protocols were elaborated on the website. | ||
| https://www.physionet.org/pn4/eegmmidb/ | ||
| ''' | ||
|
|
||
|
|
||
| import re | ||
| import requests | ||
| import pathlib | ||
| import urllib | ||
| import os | ||
|
|
||
| #%% | ||
| CONTEXT = 'pn4/' | ||
| MATERIAL = 'eegmmidb/' | ||
| URL = 'https://www.physionet.org/' + CONTEXT + MATERIAL | ||
|
|
||
| USERDIR = './data/' # Change this directory according to your setting | ||
|
|
||
| page = requests.get(URL).text | ||
| FOLDERS = sorted(list(set(re.findall(r'S[0-9]+', page)))) | ||
|
|
||
| URLS = [URL+x+'/' for x in FOLDERS] | ||
|
|
||
| # Warning: Executing this block will create folders | ||
| for folder in FOLDERS: | ||
| pathlib.Path(USERDIR +'/'+ folder).mkdir(parents=True, exist_ok=True) | ||
|
|
||
| # Warning: Executing this block will start downloading data | ||
| for i, folder in enumerate(FOLDERS): | ||
| page = requests.get(URLS[i]).text | ||
| subs = list(set(re.findall(r'S[0-9]+R[0-9]+', page))) | ||
|
|
||
| print('Working on {}, {:.1%} completed'.format(folder, (i+1)/len(FOLDERS))) | ||
| for sub in subs: | ||
| import re | ||
| import requests | ||
| import pathlib | ||
| import urllib | ||
| import os | ||
|
|
||
| #%% | ||
| CONTEXT = 'pn4/' | ||
| MATERIAL = 'eegmmidb/' | ||
| URL = 'https://www.physionet.org/' + CONTEXT + MATERIAL | ||
|
|
||
| USERDIR = './data/' # Change this directory according to your setting | ||
|
|
||
| page = requests.get(URL).text | ||
| FOLDERS = sorted(list(set(re.findall(r'S[0-9]+', page)))) | ||
|
|
||
| URLS = [URL+x+'/' for x in FOLDERS] | ||
|
|
||
| # Warning: Executing this block will create folders | ||
| for folder in FOLDERS: | ||
| pathlib.Path(USERDIR +'/'+ folder).mkdir(parents=True, exist_ok=True) | ||
|
|
||
| # Warning: Executing this block will start downloading data | ||
| for i, folder in enumerate(FOLDERS): | ||
| page = requests.get(URLS[i]).text | ||
| subs = list(set(re.findall(r'S[0-9]+R[0-9]+', page))) | ||
|
|
||
| print('Working on {}, {:.1%} completed'.format(folder, (i+1)/len(FOLDERS))) | ||
| for sub in subs: | ||
| urllib.request.urlretrieve(URLS[i]+sub+'.edf', os.path.join(USERDIR, folder, sub+'.edf')) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,149 @@ | ||
| ''' | ||
| Name: eeg_eval.py | ||
| Author: Jim Chung | ||
| Description: | ||
| This script is written to monitor the performance of neural network trained | ||
| on PhysioNet EEG data. Please check out eeg_main.py or eeg_import_py for | ||
| further information. | ||
| 'hitory.pkl' file requied in './history/' folder. | ||
| ''' | ||
|
|
||
| # load in libraries | ||
| import pickle | ||
| import matplotlib.pyplot as plt | ||
| import numpy as np | ||
| import itertools | ||
| from sklearn.metrics import confusion_matrix, classification_report, accuracy_score | ||
| from keras import models | ||
| from eeg_import import get_data, FNAMES | ||
| from eeg_preprocessing import prepare_data | ||
| import os | ||
|
|
||
| # make directories | ||
| if not os.path.exists('./metrics/'): | ||
| os.makedirs('./metrics/') | ||
|
|
||
| # functions defined | ||
|
|
||
|
|
||
| def plot_history(history): | ||
| loss_list = [s for s in history.keys() if 'loss' in s and 'val' not in s] | ||
| val_loss_list = [s for s in history.keys() if 'loss' in s and 'val' in s] | ||
| acc_list = [s for s in history.keys() if 'acc' in s and 'val' not in s] | ||
| val_acc_list = [s for s in history.keys() if 'acc' in s and 'val' in s] | ||
|
|
||
| if len(loss_list) == 0: | ||
| print('Loss is missing in history') | ||
| return | ||
|
|
||
| ## As loss always exists | ||
| epochs = range(1,len(history[loss_list[0]]) + 1) | ||
|
|
||
| ## Loss | ||
| plt.figure(1) | ||
| for l in loss_list: | ||
| plt.plot(epochs, history[l], 'b', label='Training loss (' + str(str(format(history[l][-1],'.5f'))+')')) | ||
| for l in val_loss_list: | ||
| plt.plot(epochs, history[l], 'g', label='Validation loss (' + str(str(format(history[l][-1],'.5f'))+')')) | ||
|
|
||
| plt.title('Loss') | ||
| plt.xlabel('Epochs') | ||
| plt.ylabel('Loss') | ||
| plt.legend() | ||
| plt.savefig("./metrics/loss.png") | ||
|
|
||
| ## Accuracy | ||
| plt.figure(2) | ||
| for l in acc_list: | ||
| plt.plot(epochs, history[l], 'b', label='Training accuracy (' + str(format(history[l][-1],'.5f'))+')') | ||
| for l in val_acc_list: | ||
| plt.plot(epochs, history[l], 'g', label='Validation accuracy (' + str(format(history[l][-1],'.5f'))+')') | ||
|
|
||
| plt.title('Accuracy') | ||
| plt.xlabel('Epochs') | ||
| plt.ylabel('Accuracy') | ||
| plt.legend() | ||
| plt.show() | ||
| plt.savefig("./metrics/acc.png") | ||
|
|
||
| def plot_confusion_matrix(cm, classes, | ||
| normalize=False, | ||
| cmap=plt.cm.Blues): | ||
| """ | ||
| This function prints and plots the confusion matrix. | ||
| Normalization can be applied by setting `normalize=True`. | ||
| """ | ||
| if normalize: | ||
| cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis] | ||
| title='Normalized confusion matrix' | ||
| else: | ||
| title='Confusion matrix' | ||
|
|
||
| plt.figure(3) | ||
| plt.imshow(cm, interpolation='nearest', cmap=cmap) | ||
| plt.title(title) | ||
| plt.colorbar() | ||
| tick_marks = np.arange(len(classes)) | ||
| plt.xticks(tick_marks, classes, rotation=45) | ||
| plt.yticks(tick_marks, classes) | ||
|
|
||
| fmt = '.2f' if normalize else 'd' | ||
| thresh = cm.max() / 2. | ||
| for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])): | ||
| plt.text(j, i, format(cm[i, j], fmt), | ||
| horizontalalignment="center", | ||
| color="white" if cm[i, j] > thresh else "black") | ||
|
|
||
| plt.tight_layout() | ||
| plt.ylabel('True label') | ||
| plt.xlabel('Predicted label') | ||
| plt.savefig("./metrics/confuMat.png") | ||
| plt.show() | ||
|
|
||
| def full_multiclass_report(model, | ||
| x, | ||
| y_true, | ||
| classes): | ||
|
|
||
| # 2. Predict classes and stores in y_pred | ||
| y_pred = model.predict(x).argmax(axis=1) | ||
|
|
||
| # 3. Print accuracy score | ||
| print("Accuracy : "+ str(accuracy_score(y_true,y_pred))) | ||
|
|
||
| print("") | ||
|
|
||
| # 4. Print classification report | ||
| print("Classification Report") | ||
| print(classification_report(y_true,y_pred,digits=4)) | ||
|
|
||
| # 5. Plot confusion matrix | ||
| cnf_matrix = confusion_matrix(y_true,y_pred) | ||
| print(cnf_matrix) | ||
| plot_confusion_matrix(cnf_matrix,classes=classes) | ||
|
|
||
|
|
||
| # Load in the data | ||
| howManyTest = 10 | ||
| this = np.random.randint(1, 100, size=howManyTest) | ||
| X,y = get_data([FNAMES[i] for i in this], epoch_sec=0.0625) | ||
| X_train, y_train, X_test, y_test = prepare_data(X, y) | ||
|
|
||
| print(X.shape) | ||
| print(y.shape) | ||
|
|
||
| # Get the model | ||
| model = models.load_model('./model/model0.h5') | ||
|
|
||
| # Get the history | ||
| with open('./history/history0.pkl', 'rb') as hist: | ||
| history = pickle.load(hist) | ||
|
|
||
| # Get the graphics | ||
| plot_history(history) | ||
| X_test = X_test.reshape(X_test.shape[0], X_train.shape[1], X_train.shape[2], X_train.shape[3], 1) | ||
| full_multiclass_report(model, | ||
| X_test, | ||
| y_test.argmax(axis=1), | ||
| [1,2,3,4,5]) |
Oops, something went wrong.