search_keras.py

import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.model_selection import RandomizedSearchCV
import tensorflow as tf
from tensorflow import keras
from skopt import BayesSearchCV


# In[ ]:
data = pd.read_feather('../../data_ugi/data10.feather')


# In[ ]:
y = data['READ30']
X = data.drop(['READ30'], axis=1)
X_train, X_valid, y_train, y_valid = train_test_split(X, y, train_size=0.8, test_size=0.2, random_state=1)


# In[ ]:
input_shape = [X_train.shape[1]]

def build_model(n_hidden=1, n_neurons=100, dropout=0.4, activation = "relu", learning_rate=3e-3):
    model = keras.models.Sequential()
    model.add(keras.layers.Flatten(input_shape=input_shape))
    model.add(keras.layers.BatchNormalization())
    for layer in range(n_hidden):
        model.add(keras.layers.Dense(n_neurons))
        model.add(keras.layers.BatchNormalization())
        model.add(keras.layers.Dropout(dropout))
        model.add(keras.layers.Activation(activation))
    model.add(keras.layers.Dense(1, activation="sigmoid"))

    optimizer = keras.optimizers.Adam(learning_rate=learning_rate)
    
    model.compile(loss="binary_crossentropy", metrics=['AUC'], optimizer=optimizer)
    return model


# In[ ]:


keras_clf = tf.keras.wrappers.scikit_learn.KerasClassifier(build_model)


# In[ ]:


param_distribs = {
    "n_hidden": [1, 2, 3, 4],
    "n_neurons": [25, 50, 200, 500, 1000, 1500],
    "dropout": [0.2, 0.4, 0.6, 0.8],
    "activation": ["relu", "elu"],
    "learning_rate": [3e-5, 3e-4, 3e-3, 3e-2],
}


# In[ ]:
early_stopping = keras.callbacks.EarlyStopping(
    patience=15,
    min_delta=1e-6,
    restore_best_weights=True,)

rnd_search_cv = RandomizedSearchCV(keras_clf, param_distribs, n_iter=100, scoring='roc_auc', cv=5, n_jobs=-1, verbose=2)

rnd_search_cv.fit(X_train, y_train, epochs=100, batch_size=512,
                  validation_data=(X_valid, y_valid),
                  callbacks=[early_stopping])


results = pd.DataFrame(rnd_search_cv.cv_results_)

results.sort_values(by='rank_test_score').to_csv('../../results_ugi/results_keras3_data10.csv')