3-2 Sigmoid 구현

The Graphs of Odds and Logit¶

In [12]:

import numpy as np
import tensorflow as tf

import matplotlib.pyplot as plt
plt.style.use('_classic_test_patch')

p_np = np.linspace(0.01, 0.99, 100)
p_tf = tf.linspace(0.01, 0.99, 100)

# 오즈 계산
odds_np = p_np/(1 - p_np)
odds_tf = p_tf/(1 - p_tf)

# 로짓
logit_np = np.log(odds_np)
logit_tf = tf.math.log(odds_tf)

fig, axes = plt.subplots(2, 1, figsize=(15, 10),
                         sharex=True)

axes[0].plot(p_np, odds_tf.numpy())
axes[1].plot(p_np, logit_tf.numpy())

xticks = np.arange(0, 1.1, 0.1)

axes[0].tick_params(labelsize=15)
axes[0].set_xticks(xticks)
axes[0].set_ylabel('Odds', fontsize=20, color='darkblue')
axes[1].tick_params(labelsize=15)
axes[1].set_xticks(xticks)
axes[1].set_ylabel('Logit', fontsize=20, color='darkblue')
axes[1].set_xlabel('Probability', fontsize=20, color='darkblue')

Out[12]:

Text(0.5, 0, 'Probability')

The Graphs of Sigmoid¶

In [13]:

import tensorflow as tf

from tensorflow.keras.layers import Activation

X = tf.linspace(-30, 30, 100)
sigmoid = Activation("sigmoid")(X)

fig, ax = plt.subplots(figsize=(10, 5))
ax.plot(X.numpy(), sigmoid.numpy())

Out[13]:

[<matplotlib.lines.Line2D at 0x1274274e690>]

Single-variate Logistic Regression Models¶

In [19]:

import matplotlib.pyplot as plt

import tensorflow as tf
from tensorflow.keras.layers import Dense

plt.style.use("_classic_test_patch")

X = tf.random.normal(shape=(100, 1))
dense = Dense(units=1, activation="sigmoid")

Y = dense(X)
print(Y.shape)

fig, ax = plt.subplots(figsize=(7, 7))
ax.scatter(X.numpy().flatten(), Y.numpy().flatten())

(100, 1)

Out[19]:

<matplotlib.collections.PathCollection at 0x127420838f0>

Multi-variate Logistic Regression Models¶

In [21]:

import matplotlib.pyplot as plt

import tensorflow as tf
from tensorflow.keras.layers import Dense

plt.style.use("_classic_test_patch")

X = tf.random.normal(shape=(100, 5))
dense = Dense(units=1, activation="sigmoid")

Y = dense(X)
print(Y.shape)

(100, 1)

Binary Classifier with Dense Layers¶

In [22]:

import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense

model = Sequential()
model.add(Dense(units=10, activation="relu"))
model.add(Dense(units=5, activation="relu"))
model.add(Dense(units=1, activation="sigmoid"))

In [ ]: