def fit(
self,
dataset: DatasetH,
num_boost_round=1000,
early_stopping_rounds=50,
verbose_eval=20,
evals_result=dict(),
reweighter=None,
**kwargs
):
df_train, df_valid = dataset.prepare(
["train", "valid"],
col_set=["feature", "label"],
data_key=DataHandlerLP.DK_L,
)
if df_train.empty or df_valid.empty:
raise ValueError("Empty data from dataset, please check your dataset config.")
x_train, y_train = df_train["feature"], df_train["label"]
x_valid, y_valid = df_valid["feature"], df_valid["label"]
# CatBoost needs 1D array as its label
if y_train.values.ndim == 2 and y_train.values.shape[1] == 1:
y_train_1d, y_valid_1d = np.squeeze(y_train.values), np.squeeze(y_valid.values)
else:
raise ValueError("CatBoost doesn't support multi-label training")
if reweighter is None:
w_train = None
w_valid = None
elif isinstance(reweighter, Reweighter):
w_train = reweighter.reweight(df_train).values
w_valid = reweighter.reweight(df_valid).values
else:
raise ValueError("Unsupported reweighter type.")
train_pool = Pool(data=x_train, label=y_train_1d, weight=w_train)
valid_pool = Pool(data=x_valid, label=y_valid_1d, weight=w_valid)
# Initialize the catboost model
self._params["iterations"] = num_boost_round
self._params["early_stopping_rounds"] = early_stopping_rounds
self._params["verbose_eval"] = verbose_eval
self._params["task_type"] = "GPU" if get_gpu_device_count() > 0 else "CPU"
self.model = CatBoost(self._params, **kwargs)
# train the model
self.model.fit(train_pool, eval_set=valid_pool, use_best_model=True, **kwargs)
evals_result = self.model.get_evals_result()
evals_result["train"] = list(evals_result["learn"].values())[0]
evals_result["valid"] = list(evals_result["validation"].values())[0]
class CatBoost:
_verbose = 200
_train_dir = DATA_CACHE_DIR
_is_gpu_available = get_gpu_device_count()
_task_type = "GPU" if _is_gpu_available > 0 else None
_devices = "GPU" if _is_gpu_available > 0 else None
def __init__(self, model_id, num_input_features, num_output_classes,
model_save_path, **aux_params):
self.model = CatBoostClassifier(loss_function="MultiClass",
task_type=self._task_type,
devices=self._devices,
train_dir=self._train_dir,
random_seed=SEED)
self.model.set_params(**aux_params)
self.model_id = model_id
path = f"{model_save_path}/{model_id}"
os.makedirs(path, exist_ok=True)
self.model_path = path
self.modelfile_save_path = os.path.join(path, STANDARD_MODEL_NAME)
def load(self):
self.model.load_model(self.modelfile_save_path)
def save(self):
self.model.save_model(self.modelfile_save_path)
def fit(self, X_train, y_train, X_valid, y_valid):
self.model.fit(Pool(X_train, y_train),
eval_set=(X_valid, y_valid),
use_best_model=True,
verbose=self._verbose)
self.save()
def predict(self, X, load=False):
if load:
self.load()
return self.model.predict_proba(X)
def explain(self, X_train, y_train, features, classes):
importances = self.model.get_feature_importance(
data=Pool(X_train, y_train))
plot_importance(importances, features, self.model_path, self.model_id)
def fit(self,
dataset: DatasetH,
num_boost_round=1000,
early_stopping_rounds=50,
verbose_eval=20,
evals_result=dict(),
**kwargs):
df_train, df_valid = dataset.prepare(
["train", "valid"],
col_set=["feature", "label"],
data_key=DataHandlerLP.DK_L,
)
x_train, y_train = df_train["feature"], df_train["label"]
x_valid, y_valid = df_valid["feature"], df_valid["label"]
# CatBoost needs 1D array as its label
if y_train.values.ndim == 2 and y_train.values.shape[1] == 1:
y_train_1d, y_valid_1d = np.squeeze(y_train.values), np.squeeze(
y_valid.values)
else:
raise ValueError("CatBoost doesn't support multi-label training")
train_pool = Pool(data=x_train, label=y_train_1d)
valid_pool = Pool(data=x_valid, label=y_valid_1d)
# Initialize the catboost model
self._params["iterations"] = num_boost_round
self._params["early_stopping_rounds"] = early_stopping_rounds
self._params["verbose_eval"] = verbose_eval
self._params["task_type"] = "GPU" if get_gpu_device_count(
) > 0 else "CPU"
self.model = CatBoost(self._params, **kwargs)
# train the model
self.model.fit(train_pool,
eval_set=valid_pool,
use_best_model=True,
**kwargs)
evals_result = self.model.get_evals_result()
evals_result["train"] = list(evals_result["learn"].values())[0]
evals_result["valid"] = list(evals_result["validation"].values())[0]
# https://towardsdatascience.com/catboost-vs-light-gbm-vs-xgboost-5f93620723db
import numpy as np
from catboost.utils import get_gpu_device_count
import xlsxwriter
import pandas as pd
import numpy as np
import time
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
import catboost as cb
from sklearn.model_selection import GridSearchCV
from sklearn import metrics
#import pickle
print('I see %i GPU devices' % get_gpu_device_count())
start_time = time.time()
data = pd.read_excel(open('merged final data.xlsx', 'rb'),
sheet_name='Sheet1')
data = data.replace(['Fully paid', 'Charged Off'], [0, 1])
# data['CreditScore'] = np.log(data['CreditScore'])
data.drop(['CreditScore'], axis=1, inplace=True)
data.drop(['Avg_cur_bal'], axis=1, inplace=True) # Tot_hi_cred_lim
data.drop(['Bc_util'], axis=1, inplace=True) # Revol_util
data.drop(['Total_bc_limit'], axis=1, inplace=True) # Total_rev_hi_lim
train, test, y_train, y_test = train_test_split(data.drop(["LoanStatus"], axis=1), data["LoanStatus"],
random_state=10, test_size=0.30)
params = {'depth': [2, 6],
return os.path.join(workdir, f'{filename}-{today}.{fileext}')
# 測試資料比例
# test_size_list = [ .25 ]
test_size_list = [
.05, .1, .15, .2, .25, .3, .35, .4, .45, .5, .55, .6, .65, .7, .75, .80,
.85, .90, .95
]
# 隨機種子
random_seed_list = [3, 106, 2019]
# random_seed_list = [3]
# 檢查是否有 GPU 支援
task_type = "GPU" if get_gpu_device_count() > 0 else "CPU"
model_list = [
BernoulliNB(),
CatBoostClassifier(task_type=task_type),
DecisionTreeClassifier(),
ExtraTreeClassifier(),
ExtraTreesClassifier(n_jobs=-1),
GaussianNB(),
GradientBoostingClassifier(),
GaussianProcessClassifier(multi_class='one_vs_rest',
n_jobs=-1), # Out of memory
KNeighborsClassifier(n_jobs=-1),
LGBMClassifier(n_jobs=-1),
# LabelSpreading(n_jobs = -1), # Out of Memory
LinearDiscriminantAnalysis(),
import numpy as np
from catboost.utils import get_gpu_device_count
params = {
'loss_function': 'Logloss',
'eval_metric': 'AUC',
'thread_count': -1,
'custom_metric': ['AUC:hints=skip_train~false', 'F1'],
'task_type': 'CPU' if get_gpu_device_count() > 0 else 'CPU',
# 'task_type': 'GPU', # if torch.cuda.is_available() else 'CPU',
'grow_policy': 'Lossguide', # 'SymmetricTree', # 'Depthwise',
# 'auto_class_weights': 'Balanced',
'langevin': True, # CPU only
'iterations': 20,
'learning_rate': 0.002, # 4e-3,
'l2_leaf_reg': 1e-1,
'depth': 16,
'max_leaves': 10,
'border_count': 128,
'verbose': 1,
'od_type': 'Iter',
'od_wait': 100,
# 'early_stopping_rounds': 100,
# random control
'bootstrap_type': 'Bayesian',
# 'random_seed': 100,
'random_strength': 0.001,
'rsm': 1,
'bagging_temperature': 0,
'boosting_type': 'Plain', # 'Ordered'