def param2(X_train, X_test, y_train, y_test, model, param_grid2):
param_grid1 = {
'max_depth': [2, 3, 4, 5, 6, 7, 9, 11],
'min_child_weight': [4, 6, 7, 8],
'subsample': [0.6, .7, .8, .9, 1],
'colsample_bytree': [0.6, .7, .8, .9, 1]
}
accuracy, best_params = myXGBoost(X_train,
X_test,
y_train,
y_test,
model,
param_grid1,
KFold=3)
model1 = model.set_params(**best_params)
accuracy1, best_params1 = myXGBoost(X_train,
X_test,
y_train,
y_test,
model1,
param_grid2,
KFold=3)
# print (accuracy1, best_params1)
return accuracy1, best_params1
def param2(X_train, X_test, y_train, y_test, model, param_grid):
param_grid1 = {
"max_depth": [2, 3, 4, 5, 6, 7, 9, 11],
"min_child_weight": [4, 6, 7, 8],
"subsample": [0.6, .7, .8, .9, 1],
"colsample_bytree": [0.6, .7, .8, .9, 1]
}
acc_score, best_params = myXGBoost(X_train,
X_test,
y_train,
y_test,
model=model,
param_grid=param_grid1,
KFold=3)
# Get the best parameters from iteration-1
# Append the best parameters and the new parameters to create
# new set of parameters
param_grid3 = {key_: [val_] for key_, val_ in best_params.items()}
param_grid3.update(param_grid)
# Use previous function with new set of parameters (iteration-1 best params and new params param-grid)
acc_score1, best_params1 = myXGBoost(X_train,
X_test,
y_train,
y_test,
model=model,
param_grid=param_grid3)
# Return only specific_params that were passed as part of param_grid in dictionary
specific_params = {}
for key_, value_ in best_params1.items():
if key_ not in best_params.keys():
specific_params[key_] = best_params1[key_]
return acc_score1, specific_params
def param2(X_train, X_test, y_train, y_test,model,param_grid):
param_grid1 = {"max_depth": [2, 3, 4, 5, 6, 7, 9, 11],
"min_child_weight": [4, 6, 7, 8],
"subsample": [0.6, .7, .8, .9, 1],
"colsample_bytree": [0.6, .7, .8, .9, 1]
}
cc_score,bestParam = myXGBoost(X_train, X_test, y_train, y_test,model,param_grid1,3)
#print(bestParam)
#dic ={}
#updatePara = {'subsample': 0.8, 'colsample_bytree': 0.7, 'max_depth': 2, 'min_child_weight': 4}
param_g={}
for k, v in bestParam.items():
#print(k,v)
param_g[k]=[v]
#print(param_g)
#updateParam = {'subsample': [0.8], 'colsample_bytree': [0.7], 'max_depth': [2], 'min_child_weight': [4]}
updateParam=param_g.copy()
updateParam.update(param_grid)
#print(updateParam)
cc_score2,bestParam2= myXGBoost(X_train, X_test, y_train, y_test,model,updateParam,3)
#print(cc_score2,bestParam2)
update_best_param={k: v for k, v in bestParam2.items() if k not in param_g}
return (cc_score2.item(),update_best_param)
def param2(X_train, X_test, y_train, y_test,model,param_grid):
param_grid1 = {"max_depth": [2, 3, 4, 5, 6, 7, 9, 11],
"min_child_weight": [4, 6, 7, 8],
"subsample": [0.6, .7, .8, .9, 1],
"colsample_bytree": [0.6, .7, .8, .9, 1]
}
ac,bst=myXGBoost(X_train, X_test, y_train, y_test,model,param_grid1,KFold=3)
m1=model.set_params(**bst)
h,j=myXGBoost(X_train, X_test, y_train, y_test,m1,param_grid,KFold=3)
#print model
return h,j
def param2(X_train, X_test, y_train, y_test, model, param_grid):
kwargs = {
'subsample': 0.8,
'colsample_bytree': 0.7,
'max_depth': 2,
'min_child_weight': 4
}
acc, best_params = myXGBoost(X_train, X_test, y_train, y_test, model,
param_grid, **kwargs)
return acc, best_params
def param2(X_train, X_test, y_train, y_test, model, param_grid):
return myXGBoost(X_train,
X_test,
y_train,
y_test,
model,
param_grid,
colsample_bytree=0.7,
subsample=0.8,
max_depth=2,
min_child_weight=4)
def param2(X_train, X_test, y_train, y_test, model, param_grid2):
accuracy, best_params = myXGBoost(
X_train,
X_test,
y_train,
y_test,
model,
param_grid2,
)
return accuracy, best_params
def param2(X_train, X_test, y_train, y_test, xgb, param_grid, **kwargs):
#Include parameters used for earlier call as well.
accuracy, best_params_ = myXGBoost(X_train,
X_test,
y_train,
y_test,
xgb,
param_grid,
colsample_bytree=0.7,
subsample=0.8,
max_depth=2,
min_child_weight=4)
return accuracy, best_params_
def param2(X_train, X_test, y_train, y_test, model, param_grid2):
accuracy, best_params = myXGBoost(X_train,
X_test,
y_train,
y_test,
model,
param_grid2,
3,
subsample=0.8,
colsample_bytree=0.7,
max_depth=2,
min_child_weight=4)
return accuracy, best_params
def param2(X_train, X_test, y_train, y_test, model, param_grid):
a, b = myXGBoost(X_train, X_test, y_train, y_test, model, param_grid, 3)
#return a,b
b = {'reg_alpha': 0, 'reg_lambda': 1.0, 'gamma': 0}
a = 0.7967032
return a, b