def run():
cols = []
for i in range(0, 54):
if i != 37:
cols.append(i)
data = np.genfromtxt(
"Features_Variant_1.csv", delimiter=",", usecols=cols
) # читка данных, константный столбец можно было из самой csv-шки удалить, но я решил не трогать
shuffle(data) # перемешиваем данные
x = data[:, 0:len(data[0]) - 1] # забираем данные без целевой переменной
y = data[:, len(data[0]) - 1] # целевая переменная
# листы с фолдами
x_list = []
y_list = []
# начальные данные
number_of_folds = 5
learning_rate = 0.0001
batch_size = 5000
initial_w = np.zeros(len(x[0]))
initial_e = 0
amount_of_iterations = 1000
train_rmse = []
test_rmse = []
train_r2 = []
test_r2 = []
for i in range(0, number_of_folds):
x_list.append(x[len(x) * i // number_of_folds:len(x) * (i + 1) //
number_of_folds, :])
y_list.append(y[len(y) * i // number_of_folds:len(y) * (i + 1) //
number_of_folds])
for i in range(0, number_of_folds):
costil = 0
if i != 0:
x_train = np.copy(x_list[0])
y_train = np.copy(y_list[0])
else:
x_train = np.copy(x_list[1])
y_train = np.copy(y_list[1])
costil = 1
x_test = np.copy(x_list[i])
y_test = np.copy(y_list[i])
for j in range(1 + costil, number_of_folds):
if j != i:
x_train = np.concatenate((x_train, x_list[j]), axis=0)
y_train = np.concatenate((y_train, y_list[j]), axis=0)
x_means, x_sds = get_means_and_sds(x_train)
x_train = normalize_data_st(x_train, x_means, x_sds)
x_test = normalize_data_st(x_test, x_means, x_sds)
[w, e] = gradient_descent_runner(x_train, y_train, initial_w,
initial_e, learning_rate,
amount_of_iterations, batch_size)
train_rmse.append(compute_rmse(w, e, x_train, y_train))
test_rmse.append(compute_rmse(w, e, x_test, y_test))
train_r2.append(compute_r2(w, e, x_train, y_train))
test_r2.append(compute_r2(w, e, x_test, y_test))
log_local_data(w, e, compute_mse(w, e, x_train, y_train),
compute_mse(w, e, x_test, y_test), i + 1, "output.txt")
log_data(train_rmse, test_rmse, train_r2, test_r2, learning_rate,
batch_size, "output.txt")
def test_DFM_avazu(data, train, test):
print("\nTesting DFM on avazu dataset...\n")
results_activation_function = {"auc": [], "logloss": [], "rmse": []}
results_dropout = {"auc": [], "logloss": [], "rmse": []}
results_number_of_neurons = {"auc": [], "logloss": [], "rmse": []}
auc = 0
logloss = 0
rmse = 0
features_labels = train.columns
sparse_features_labels = features_labels[1:23]
target_label = features_labels[0]
dnn_feature_columns = [
SparseFeat(
feat,
vocabulary_size=data[feat].nunique(),
embedding_dim=4,
) for feat in sparse_features_labels
]
linear_feature_columns = [
SparseFeat(
feat,
vocabulary_size=data[feat].nunique(),
embedding_dim=4,
) for feat in sparse_features_labels
]
feature_names = get_feature_names(linear_feature_columns +
dnn_feature_columns)
train_model_input = {name: train[name] for name in feature_names}
test_model_input = {name: test[name] for name in feature_names}
true_y = test[target_label].values
print("\t\t-- ACTIVATION FUNCTIONS --\t\t")
for dnn_activation in dnn_activation_list:
print("\nTesting {dnn_activation}...".format(
dnn_activation=dnn_activation))
model = DeepFM(linear_feature_columns,
dnn_feature_columns,
dnn_activation=dnn_activation,
task='binary')
model.compile(
"adam",
"binary_crossentropy",
metrics=['binary_crossentropy'],
)
model.fit(
train_model_input,
train[target_label].values,
batch_size=256,
epochs=10,
verbose=0,
validation_split=TEST_PROPORTION,
)
pred_y = model.predict(test_model_input, batch_size=256)
auc = compute_auc(true_y, pred_y)
logloss = compute_log_loss(true_y, pred_y)
rmse = compute_rmse(true_y, pred_y)
results_activation_function["auc"].append(auc)
results_activation_function["logloss"].append(logloss)
results_activation_function["rmse"].append(rmse)
print("\t\t-- DROPOUT RATES --\t\t")
for dnn_dropout in dnn_dropout_list:
print("\nTesting {dnn_dropout}...".format(dnn_dropout=dnn_dropout))
model = DeepFM(linear_feature_columns,
dnn_feature_columns,
dnn_dropout=dnn_dropout,
task='binary')
model.compile(
"adam",
"binary_crossentropy",
metrics=['binary_crossentropy'],
)
model.fit(
train_model_input,
train[target_label].values,
batch_size=256,
epochs=10,
verbose=0,
validation_split=TEST_PROPORTION,
)
pred_y = model.predict(test_model_input, batch_size=256)
auc = compute_auc(true_y, pred_y)
logloss = compute_log_loss(true_y, pred_y)
rmse = compute_rmse(true_y, pred_y)
results_dropout["auc"].append(auc)
results_dropout["logloss"].append(logloss)
results_dropout["rmse"].append(rmse)
print("\t\t-- HIDDEN UNITS --\t\t")
for dnn_hidden_units in dnn_hidden_units_list:
print("\nTesting {dnn_hidden_units}...".format(
dnn_hidden_units=dnn_hidden_units))
model = DeepFM(linear_feature_columns,
dnn_feature_columns,
dnn_hidden_units=dnn_hidden_units,
task='binary')
model.compile(
"adam",
"binary_crossentropy",
metrics=['binary_crossentropy'],
)
model.fit(
train_model_input,
train[target_label].values,
batch_size=256,
epochs=10,
verbose=0,
validation_split=TEST_PROPORTION,
)
pred_y = model.predict(test_model_input, batch_size=256)
auc = compute_auc(true_y, pred_y)
logloss = compute_log_loss(true_y, pred_y)
rmse = compute_rmse(true_y, pred_y)
results_number_of_neurons["auc"].append(auc)
results_number_of_neurons["logloss"].append(logloss)
results_number_of_neurons["rmse"].append(rmse)
if PLOT:
create_plots("DFM", "avazu", results_activation_function,
"Activation Function", "activation_func",
dnn_activation_list)
create_plots("DFM", "avazu", results_dropout, "Dropout Rate",
"dropout", dnn_dropout_list)
create_plots("DFM", "avazu", results_number_of_neurons,
"Number of Neurons per layer", "nr_neurons",
dnn_hidden_units_list)