aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorVasil Zlatanov <vasil@netcraft.com>2018-11-14 22:19:14 +0000
committerVasil Zlatanov <vasil@netcraft.com>2018-11-14 22:19:14 +0000
commita94130756408dcb0884e5edca00408e2faf2c65b (patch)
treeca1edc0b08f7fe35d5bade66a2835c6817f44420
parent868a6d52f99b49adb556cffc6b59c00de153d355 (diff)
downloadvz215_np1915-a94130756408dcb0884e5edca00408e2faf2c65b.tar.gz
vz215_np1915-a94130756408dcb0884e5edca00408e2faf2c65b.tar.bz2
vz215_np1915-a94130756408dcb0884e5edca00408e2faf2c65b.zip
Simplify code by removing repetition
-rwxr-xr-xtrain.py69
1 files changed, 23 insertions, 46 deletions
diff --git a/train.py b/train.py
index 349d494..ef2df18 100755
--- a/train.py
+++ b/train.py
@@ -63,18 +63,17 @@ def test_split(n_faces, raw_faces, split, seed):
faces_test = faces_test.reshape(n_faces*n_test_faces, n_pixels)
return faces_train, faces_test, target_train, target_test
-def draw_conf_mat(args, target_test, target_pred):
+def draw_results(args, target_test, target_pred):
acc_sc = accuracy_score(target_test, target_pred)
- if not args.classifyalt:
- cm = confusion_matrix(target_test, target_pred)
- print('Accuracy: ', acc_sc)
- if (args.conf_mat):
- plt.matshow(cm, cmap='Blues')
- plt.colorbar()
- plt.ylabel('Actual')
- plt.xlabel('Predicted')
- plt.show()
- return acc_sc
+ cm = confusion_matrix(target_test, target_pred)
+ print('Accuracy: ', acc_sc)
+ if (args.conf_mat):
+ plt.matshow(cm, cmap='Blues')
+ plt.colorbar()
+ plt.ylabel('Actual')
+ plt.xlabel('Predicted')
+ plt.show()
+ return
def test_model(M, faces_train, faces_test, target_train, target_test, args):
raw_faces_train = faces_train
@@ -185,8 +184,7 @@ def test_model(M, faces_train, faces_test, target_train, target_test, args):
plt.show()
#Better n_neighbors = 2
- accuracy = draw_conf_mat(args, target_test, target_pred)
- return target_pred, accuracy, distances
+ return target_pred, distances
def main():
parser = argparse.ArgumentParser()
@@ -233,32 +231,23 @@ def main():
faces_train = faces_train.reshape(n_faces, int(faces_train.shape[0]/n_faces), n_pixels)
target_train = target_train.reshape(n_faces, int(target_train.shape[0]/n_faces))
- accuracy = np.zeros(n_faces)
distances = np.zeros((n_faces, faces_test.shape[0]))
for i in range(n_faces):
- target_pred, accuracy[i], distances[i] = test_model(args.eigen, faces_train[i], faces_test, target_train[i], target_test, args)
+ target_pred, distances[i] = test_model(args.eigen, faces_train[i], faces_test, target_train[i], target_test, args)
target_pred = np.argmin(distances, axis=0)
- acc_sc = accuracy_score(target_test, target_pred)
- cm = confusion_matrix(target_test, target_pred)
- print('Total Accuracy: ', acc_sc)
- if (args.conf_mat):
- plt.matshow(cm, cmap='Blues')
- plt.colorbar()
- plt.ylabel('Actual')
- plt.xlabel('Predicted')
- plt.show()
- return
+ elif args.reigen:
+ target_pred = np.zeros((args.reigen-args.eigen, 2*n_faces))
+ accuracy = np.zeros((args.reigen-args.eigen, 2*n_faces))
+ rec_error = np.zeros((args.reigen-args.eigen, 2*n_faces))
- if args.reigen:
- accuracy = np.zeros(args.reigen-args.eigen)
- rec_error = np.zeros((args.reigen-args.eigen, 2*n_faces))
for M in range(args.eigen, args.reigen):
start = timer()
- target_pred, accuracy[M - args.eigen], rec_error[M - args.eigen] = test_model(M, faces_train, faces_test, target_train, target_test, args)
+ target_pred[i], rec_error[M - args.eigen] = test_model(M, faces_train, faces_test, target_train, target_test, args)
end = timer()
print("Run with", M, "eigenvalues completed in ", end-start, "seconds")
print("Memory Used:", psutil.Process(os.getpid()).memory_info().rss)
- #plot
+ accuracy[i] = accuracy_score(target_test, target_pred[i])
+ # Plot
print('Max efficiency of ', max(accuracy), '% for M =', np.argmax(accuracy))
plt.plot(range(args.eigen, args.reigen), 100*accuracy)
plt.xlabel('Number of Eigenvectors used (M)')
@@ -266,34 +255,22 @@ def main():
plt.grid(True)
plt.show()
elif args.ensemble:
- accuracy = np.zeros(args.ensemble)
rec_error = np.zeros((args.ensemble, n_faces, faces_test.shape[0]))
target_pred = np.zeros((args.ensemble, target_test.shape[0]))
for i in range(args.ensemble):
- target_pred[i], accuracy[i], rec_error[i] = test_model(args.eigen, faces_train_bagged[i], faces_test, target_train, target_test, args)
+ target_pred[i], rec_error[i] = test_model(args.eigen, faces_train_bagged[i], faces_test, target_train, target_test, args)
target_pred_comb = np.zeros(target_pred.shape[1])
target_pred = target_pred.astype(int).T
for i in range(target_pred.shape[0]):
target_pred_comb[i] = np.bincount(target_pred[i]).argmax()
- print(target_pred_comb)
- acc_sc = accuracy_score(target_test, target_pred_comb)
- cm = confusion_matrix(target_test, target_pred_comb)
- print('Total Accuracy: ', acc_sc)
- if (args.conf_mat):
- plt.matshow(cm, cmap='Blues')
- plt.colorbar()
- plt.ylabel('Actual')
- plt.xlabel('Predicted')
- plt.show()
- return
-
-
+ target_pred = target_pred_comb
else:
M = args.eigen
start = timer()
- test_model(M, faces_train, faces_test, target_train, target_test, args)
+ target_pred, distances = test_model(M, faces_train, faces_test, target_train, target_test, args)
end = timer()
+ draw_results(args, target_test, target_pred)
if __name__ == "__main__":
main()