Spaces:
Sleeping
Sleeping
Jensen-holm
commited on
Commit
·
9c1370a
1
Parent(s):
d6cc61a
changing the example to what is actually in the example
Browse files
README.md
CHANGED
@@ -26,7 +26,7 @@ A small, simple neural network framework built using only [numpy](https://numpy.
|
|
26 |
from sklearn import datasets
|
27 |
from sklearn.preprocessing import OneHotEncoder
|
28 |
from sklearn.model_selection import train_test_split
|
29 |
-
from sklearn.metrics import accuracy_score
|
30 |
import numpy as np
|
31 |
from numpyneuron import (
|
32 |
NN,
|
@@ -39,7 +39,7 @@ from numpyneuron import (
|
|
39 |
RANDOM_SEED = 2
|
40 |
|
41 |
|
42 |
-
def
|
43 |
seed: int,
|
44 |
) -> tuple[np.ndarray, ...]:
|
45 |
digits = datasets.load_digits(as_frame=False)
|
@@ -55,9 +55,10 @@ def _preprocess_digits(
|
|
55 |
return X_train, X_test, y_train, y_test
|
56 |
|
57 |
|
58 |
-
def train_nn_classifier(
|
59 |
-
X_train
|
60 |
-
|
|
|
61 |
nn_classifier = NN(
|
62 |
epochs=2_000,
|
63 |
hidden_size=16,
|
@@ -75,19 +76,19 @@ def train_nn_classifier() -> None:
|
|
75 |
X_train=X_train,
|
76 |
y_train=y_train,
|
77 |
)
|
|
|
|
|
78 |
|
79 |
-
|
|
|
|
|
80 |
|
|
|
81 |
pred = np.argmax(pred, axis=1)
|
82 |
y_test = np.argmax(y_test, axis=1)
|
83 |
|
84 |
accuracy = accuracy_score(y_true=y_test, y_pred=pred)
|
85 |
-
|
86 |
print(f"accuracy on validation set: {accuracy:.4f}")
|
87 |
-
|
88 |
-
|
89 |
-
if __name__ == "__main__":
|
90 |
-
train_nn_classifier()
|
91 |
```
|
92 |
|
93 |
## Running Example
|
|
|
26 |
from sklearn import datasets
|
27 |
from sklearn.preprocessing import OneHotEncoder
|
28 |
from sklearn.model_selection import train_test_split
|
29 |
+
from sklearn.metrics import accuracy_score
|
30 |
import numpy as np
|
31 |
from numpyneuron import (
|
32 |
NN,
|
|
|
39 |
RANDOM_SEED = 2
|
40 |
|
41 |
|
42 |
+
def preprocess_digits(
|
43 |
seed: int,
|
44 |
) -> tuple[np.ndarray, ...]:
|
45 |
digits = datasets.load_digits(as_frame=False)
|
|
|
55 |
return X_train, X_test, y_train, y_test
|
56 |
|
57 |
|
58 |
+
def train_nn_classifier(
|
59 |
+
X_train: np.ndarray,
|
60 |
+
y_train: np.ndarray,
|
61 |
+
) -> NN:
|
62 |
nn_classifier = NN(
|
63 |
epochs=2_000,
|
64 |
hidden_size=16,
|
|
|
76 |
X_train=X_train,
|
77 |
y_train=y_train,
|
78 |
)
|
79 |
+
return nn_classifier
|
80 |
+
|
81 |
|
82 |
+
if __name__ == "__main__":
|
83 |
+
X_train, X_test, y_train, y_test = preprocess_digits(seed=RANDOM_SEED)
|
84 |
+
classifier = train_nn_classifier(X_train, y_train)
|
85 |
|
86 |
+
pred = classifier.predict(X_test)
|
87 |
pred = np.argmax(pred, axis=1)
|
88 |
y_test = np.argmax(y_test, axis=1)
|
89 |
|
90 |
accuracy = accuracy_score(y_true=y_test, y_pred=pred)
|
|
|
91 |
print(f"accuracy on validation set: {accuracy:.4f}")
|
|
|
|
|
|
|
|
|
92 |
```
|
93 |
|
94 |
## Running Example
|