Lee y mejora este código: import pandas as pd import numpy as np from keras.models import Sequential from keras.layers import Dense, Dropout # Importar la capa Dropout from keras.callbacks import EarlyStopping from keras.regularizers import l1 # Importar el regularizador L1 from sklearn.preprocessing import StandardScaler # Leer el archivo de Excel df = pd.read_excel('/content/drive/MyDrive/Producción.xlsx', sheet_name=0, usecols='A:CG', nrows=15) # Agregar las filas para los años 2020 a 2031 new_rows = pd.DataFrame({'AÑO': range(2020, 2031)}) df = pd.concat([df, new_rows], ignore_index=True) data = df.iloc[:-11, 1:].values # Datos if np.isnan(data).any(): print("Warning: Training data contains NaN values. Removing rows with NaN values.") # Eliminar las filas con valores NaN mask = ~np.isnan(data).any(axis=1) data = data[mask] scaler = StandardScaler() data = scaler.fit_transform(data) X = data[:-1] y = data[1:] split_index = int(len(X) * 0.8) X_train, X_val = X[:split_index], X[split_index:] y_train, y_val = y[:split_index], y[split_index:] model = Sequential() model.add(Dense(128, input_dim=X.shape[1], activation='relu', activity_regularizer=l1(0.0001))) # Agregar regularización L1 a la primera capa model.add(Dropout(0.2)) # Agregar capa Dropout después de la primera capa model.add(Dense(32, activation='relu')) model.add(Dense(16, activation='relu')) model.add(Dense(y.shape[1])) model.compile(loss='mean_squared_error', optimizer='sgd') early_stopping = EarlyStopping(monitor='val_loss', patience=10) history = model.fit(X_train, y_train, validation_data=(X_val, y_val), epochs=1000, callbacks=[early_stopping]) # Obtener las últimas observaciones para cada municipio last_observations = data[-1:] # Realizar pronósticos para cada año for year in range(2020, 2031): normalized_forecasts = model.predict(last_observations) # Utilizar la red neuronal para generar los pronósticos normalizados forecasts = scaler.inverse_transform(normalized_forecasts) df.loc[df['AÑO'] == year, df.columns[1:]] = forecasts last_observations = np.append(last_observations, normalized_forecasts, axis=0)[-1:] print(df) # Guardar los resultados en un nuevo archivo de Excel df.to_excel('pron3.xlsx')

function that: import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.metrics import classification_report, confusion_matrix
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import accuracy_score
from sklearn.model_selection import RandomizedSearchCV
from sklearn.metrics import plot_confusion_matrix
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import cross_val_score

# Leer el archivo de Excel
df = pd.read_excel('/content/drive/MyDrive/data_depurada.xlsx', sheet_name=0, usecols='A:M', nrows=15)


# Seleccionar las columnas (variables) que se usarán en el entrenamiento
data = df.iloc[:, 2:].values
X = data[:, :-1] # Seleccionar todas las filas