lenoctambule/autoencoder
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: bad cmp for patience in train_dataset method

Browse Source
This commit is contained in:
Lenoctambule
2026-03-27 07:34:04 +01:00
parent 9859863ec9
commit 3301588f72
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
autoencoder.py
View File

@@ -53,8 +53,7 @@ class Decoder:
self.last_output = regularize(self.activation_func(res)) self.last_output = regularize(self.activation_func(res))
return self.last_output return self.last_output
def backprop(self, target: np.ndarray): def backprop(self, error: np.ndarray):
error = self.last_output - target
dW = np.outer(self.last_input, error) dW = np.outer(self.last_input, error)
self.W -= self.lr * dW self.W -= self.lr * dW
self.B -= self.lr * error self.B -= self.lr * error
@@ -73,7 +72,7 @@ class Autoencoder:
def train(self, v: np.ndarray) -> float: def train(self, v: np.ndarray) -> float:
encoded = self.encoder.forward(v) encoded = self.encoder.forward(v)
reconstructed = self.decoder.forward(encoded) reconstructed = self.decoder.forward(encoded)
error = self.decoder.backprop(v) error = self.decoder.backprop(v - reconstructed)
self.encoder.backprop(error) self.encoder.backprop(error)
error = v - reconstructed error = v - reconstructed
return np.sum(np.abs(error)) return np.sum(np.abs(error))
@@ -99,7 +98,7 @@ class Autoencoder:
input = x.flatten() input = x.flatten()
error += self.train(input) error += self.train(input)
error /= len(data_set) error /= len(data_set)
if error - prev_error <= 1e-8: if prev_error - error <= 1e-8:
no_improv += 1 no_improv += 1
else: else:
no_improv = 0 no_improv = 0
@@ -114,6 +113,7 @@ class Autoencoder:
epoch += 1 epoch += 1
if display_loss is True: if display_loss is True:
dynamic_loss_plot_finish(ax, line) dynamic_loss_plot_finish(ax, line)
print("\r#Training complete !")
return losses return losses
def encode(self, v: np.ndarray) -> np.ndarray: def encode(self, v: np.ndarray) -> np.ndarray: