Readme updated sklearn

Readme.md

@@ -13,7 +13,7 @@ celsius = np.array([-40, -10, 0, 8, 15, 22, 38], dtype=float)
 fahrenheit = np.array([-40, 14, 32, 46, 59, 72, 100], dtype=float)
 
 # option 2: (X°C x 9/5) + 32 = 41 °F
-points = 100
+points = 1000
 np.random.seed(99)
 dataIn = np.linspace (-40,60, points)
 target = dataIn*9/5 + 32 +4*np.random.randn(points)
@@ -94,8 +94,7 @@ from tensorflow.keras.optimizers import Adam
 #hyper parameters
 epoch = 500
 lr = 0.01
-hn = 2 # hidden nodes
-tf.random.set_seed(42)  # For TensorFlow
+tf.random.set_seed(99)  # For TensorFlow
 
 
 model.compile(optimizer=Adam(lr), loss='mean_squared_error')
@@ -109,25 +108,50 @@ print("Model trainned!")
 
 
 
+```python
+plt.plot(historial.epoch, historial.history['loss'], '.k' )
+plt.show()
+```
+
+
+    
+![png](Readme_files/Readme_5_0.png)
+    
+
+
+
 ```python
 predict = model.predict(dataIn)
-plt.plot(dataIn, predict, ':r', label='estimated')
-plt.plot(dataIn,target, '.b', label='real', alpha=0.4)
+plt.plot(dataIn, predict, '-r', label='estimated')
+plt.plot(dataIn,target, '.b', label='real', alpha=0.1)
+#plt.xlim([0, 20])
+#plt.ylim([32, 39])
 plt.legend()
 plt.grid()
 plt.show()
 ```
 
-    4/4 ━━━━━━━━━━━━━━━━━━━━ 0s 4ms/step 
+    32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 741us/step
 
 
 
     
-![png](Readme_files/Readme_5_1.png)
+![png](Readme_files/Readme_6_1.png)
     
 
 
 
+```python
+for layer in model.layers:
+    print(layer.get_weights())
+```
+
+    [array([[ 0.7760521 , -0.18955402]], dtype=float32), array([8.428659, 8.034532], dtype=float32)]
+    [array([[2.4613197 ],
+           [0.63733613]], dtype=float32), array([6.2560296], dtype=float32)]
+
+
+
 ```python
 # Get weights
 for layer in model.layers:
@@ -139,15 +163,49 @@ for layer in model.layers:
 
     Layer: hidden
       Weights (Kernel): (1, 2) 
-    [[-0.27738443  0.7908125 ]]
+    [[ 0.7760521  -0.18955402]]
       Biases: (2,) 
-    [-8.219968  6.714554]
+    [8.428659 8.034532]
     Layer: output
       Weights (Kernel): (2, 1) 
-    [[-1.9934888]
-     [ 1.5958738]]
+    [[2.4613197 ]
+     [0.63733613]]
       Biases: (1,) 
-    [5.1361823]
+    [6.2560296]
+
+
+
+```python
+inData = np.array([100])
+model.predict(inData)
+```
+
+    1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 25ms/step
+
+
+
+
+
+    array([[211.05261]], dtype=float32)
+
+
+
+
+```python
+wih = np.array([[ 0.7760521,  -0.18955402]])
+bh = np.array([8.428659, 8.034532])
+Xh = np.dot(inData, wih) + bh
+who = np.array([[2.4613197 ],[0.63733613]])
+bo = np.array([6.2560296])
+O = np.dot(Xh,who) + bo
+O
+```
+
+
+
+
+    array([211.05262121])
+
 
 
 # Testing the model
@@ -158,13 +216,13 @@ inTest = np.array([100])
 model.predict(inTest)
 ```
 
-    1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 87ms/step
+    1/1 ━━━━━━━━━━━━━━━━━━━━ 0s 22ms/step
 
 
 
 
 
-    array([[213.73816]], dtype=float32)
+    array([[115.929985]], dtype=float32)
 
 
 
@@ -188,6 +246,91 @@ Oo
 
 
 
+# sklearn
+
+
+```python
+from sklearn.neural_network import MLPRegressor
+from sklearn.preprocessing import StandardScaler
+from sklearn.model_selection import train_test_split
+import numpy as np
+
+# Datos de ejemplo
+
+
+# Escalado de los datos
+scaler_X = StandardScaler()
+scaler_y = StandardScaler()
+
+X_scaled = scaler_X.fit_transform(dataIn.reshape(-1,1))
+y_scaled = scaler_y.fit_transform(target.reshape(-1, 1)).ravel()
+
+# Modelo equivalente al de Keras
+mlp = MLPRegressor(
+    hidden_layer_sizes=(2,),     # 1 capa oculta con 2 neuronas
+    activation='identity',       # activación lineal
+    learning_rate_init=0.001,  # 👈 Learning rate
+    solver='adam',
+    max_iter=1000,
+    tol=1e-6,
+    random_state=4
+)
+
+# Entrenar modelo
+mlp.fit(X_scaled, y_scaled)
+
+# Predicción
+y_pred_scaled = mlp.predict(X_scaled)
+y_pred = scaler_y.inverse_transform(y_pred_scaled.reshape(-1, 1))
+
+# Visualizar resultados (opcional)
+import matplotlib.pyplot as plt
+
+plt.scatter(dataIn, target, label="Original data")
+plt.plot(dataIn, y_pred, color='red', label="MLPRegressor output")
+plt.legend()
+plt.show()
+```
+
+
+    
+![png](Readme_files/Readme_15_0.png)
+    
+
+
+
+```python
+plt.plot(mlp.loss_curve_,'.k')
+plt.xlabel("Épocas")
+plt.ylabel("Error (loss)")
+plt.title("Evolución del error en entrenamiento")
+plt.grid(True)
+plt.show()
+```
+
+
+    
+![png](Readme_files/Readme_16_0.png)
+    
+
+
+
+```python
+print("Pesos entre capa de entrada y oculta:", mlp.coefs_[0])
+print("Pesos entre capa oculta y salida:", mlp.coefs_[1])
+print("Bias de capa oculta:", mlp.intercepts_[0])
+print("Bias de salida:", mlp.intercepts_[1])
+```
+
+    Pesos entre capa de entrada y oculta: [[ 1.70549238 -0.37235861]]
+    Pesos entre capa oculta y salida: [[ 0.30934654]
+     [-1.25842791]]
+    Bias de capa oculta: [1.02819949 1.02732046]
+    Bias de salida: [0.97683886]
+
+
+# Model scheme
+
 
 ```python
 def generate_ascii_ann(model):