2211275
/
gnn
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

new repo structure

main
romanamo 2024-05-09 12:28:50 +02:00
parent 176ed663c1
commit e6e35776b8
6 changed files with 10036 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.gitignore vendored
View File

@ -172,5 +172,5 @@ cython_debug/
# be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore # be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
# and can be added to the global gitignore or merged into this file. For a more nuclear # and can be added to the global gitignore or merged into this file. For a more nuclear
# option (not recommended) you can uncomment the following to ignore the entire idea folder. # option (not recommended) you can uncomment the following to ignore the entire idea folder.
#.idea/ .idea/

12
README.md
View File

@ -1,3 +1,15 @@
# gnn # gnn
Code-Snippets für die Vorlesung "Grundlagen Neuronale Netze" im SS24. Code-Snippets für die Vorlesung "Grundlagen Neuronale Netze" im SS24.
## Aufgaben
| Nr | Aufgabenstellung | Kommentar |
|----|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------|
| 1 | Verändern Sie die Lernregel so, dass es die Perzeptron Lernregel ist. | |
| 2 | Verändern Sie den Backpropagation Algorithmus so, dass er den Gradientenabstieg vollständig mit iRProp− macht. | |
| 3 | Implementieren Sie eine Restricted Boltzmann Machine als einfachste Form eines Autoencoders. Am Eingang sollen die MNIST Ziffern angelegt werden. Durch Aktivierung der Hidden Schicht und Rückaktivierung sollten die MNIST Ziffern wieder rekonstruiert werden können. Nun reduzieren Sie die Anzahl der Hidden Neuronen auf 100. Können die Ziffern trotzdem so rekonstruiert werden, dass man sie noch erkennen kann? | |
| 4 | | |
| 5 | | |
| 6 | | |

0
uebungen/uebung1.py → uebungen/aufgabe1/uebung1.py
View File

0
uebungen/uebung2.py → uebungen/aufgabe2/uebung2.py
View File

10000
uebungen/aufgabe3/mnist.csv 100644
View File

File diff suppressed because it is too large Load Diff

43
uebungen/uebung3.py → uebungen/aufgabe3/uebung3.py
View File

@ -4,7 +4,7 @@ import matplotlib.pyplot as plt
def load_data(): def load_data():
df_orig_train = pd.read_csv('mnist_test_final.csv') df_orig_train = pd.read_csv('mnist.csv')
df_digits = df_orig_train.drop('label', axis=1) df_digits = df_orig_train.drop('label', axis=1)
return df_digits.to_numpy() return df_digits.to_numpy()
@ -19,8 +19,8 @@ def sigmoid(x):
class RBM: class RBM:
def __init__(self, visible_size: int, hidden_size: int, learnrate: float = 0.1, epochs=20): def __init__(self, visible_size: int, hidden_size: int, learnrate: float = 0.1, epochs: int = 20):
"""__init__ Initializes a newly created Ristricted Bolzmann Machine. """__init__ Initializes a newly created Restricted Bolzmann Machine.
Args: Args:
visible_size (int): amount of neurons inside the visible layer visible_size (int): amount of neurons inside the visible layer
@ -33,9 +33,6 @@ class RBM:
self.hidden_size = hidden_size self.hidden_size = hidden_size
self.epochs = epochs self.epochs = epochs
self.reset()
def reset(self):
# initialize/reset learnable attributes # initialize/reset learnable attributes
self.weights = np.random.randn(self.visible_size, self.hidden_size) self.weights = np.random.randn(self.visible_size, self.hidden_size)
self.visible_bias = np.zeros(self.visible_size) * 0.1 self.visible_bias = np.zeros(self.visible_size) * 0.1
@ -80,34 +77,30 @@ class RBM:
return h0, v1 return h0, v1
rbm = RBM(28**2, 256, 0.1, epochs=3)
def validate(idx): rbm = RBM(28 ** 2, 100, 0.2, epochs=1)
#flatten and normalize mnist data
test = mnist[idx].flatten()/255
# train bolzmann machine and run for i in range(100):
rbm.train(test) # normalize mnist data and train
(hid, out) = rbm.run(test) number = mnist[i] / 255
rbm.train(number)
return (hid.reshape((16, 16)), out.reshape((28, 28)))
# plot results # plot results
rows, columns = (4, 6) rows, columns = (9, 9)
fig = plt.figure(figsize=(10, 7)) fig = plt.figure(figsize=(10, 7))
fig.canvas.manager.set_window_title("Reconstruction of MNIST Numbers using a Restricted Boltzmann Machine")
for i in range((rows * columns)): for i in range((rows * columns)):
if i % 2 == 0: if i % 3 == 0:
(hid, out) = validate(i) number = mnist[i] / 255
(hidden, visible) = rbm.run(number)
# hidden layer results = [hidden.reshape((10, 10)), visible.reshape((28, 28)), number.reshape((28, 28))]
fig.add_subplot(rows, columns, i+1)
plt.imshow(hid, cmap='gray')
plt.axis('off')
# visible layer for j, item in enumerate(results):
fig.add_subplot(rows, columns, i+2) fig.add_subplot(rows, columns, i + j + 1)
plt.imshow(out, cmap='gray') plt.imshow(item, cmap='gray')
plt.axis('off') plt.axis('off')
plt.show() plt.show()