From ed422d7e57cdbab0c246e3660de01034bbbaad4d Mon Sep 17 00:00:00 2001
From: Li Hengyu <66453357+HengyuLi-Ozaki-lab@users.noreply.github.com>
Date: Tue, 3 Oct 2023 11:27:57 +0900
Subject: [PATCH 1/5] Add GP_sfs class for satisfying scikit-learn

---
 physbo/gp/core/model.py | 76 +++++++++++++++++++++++++++++++++++++++--
 1 file changed, 73 insertions(+), 3 deletions(-)

diff --git a/physbo/gp/core/model.py b/physbo/gp/core/model.py
index 0bdd51a7..64b692ef 100644
--- a/physbo/gp/core/model.py
+++ b/physbo/gp/core/model.py
@@ -4,10 +4,10 @@
 from physbo.gp import inf
 from physbo.gp.core import learning
 from physbo.gp.core.prior import prior
-
+from physbo.misc import set_config
 
 class model:
-    def __init__(self, lik, mean, cov, inf="exact"):
+    def __init__(self, lik, mean, cov, xtrain=None, ytrain=None, inf="exact"):
         """
 
         Parameters
@@ -20,6 +20,8 @@ def __init__(self, lik, mean, cov, inf="exact"):
         self.lik = lik
         self.prior = prior(mean=mean, cov=cov)
         self.inf = inf
+        self.xtrain = xtrain
+        self.ytrain = ytrain
         self.num_params = self.lik.num_params + self.prior.num_params
         self.params = self.cat_params(self.lik.params, self.prior.params)
         self.stats = ()
@@ -301,7 +303,7 @@ def post_sampling(self, X, Z, params=None, N=1, alpha=1):
 
         fmean = self.get_post_fmean(X, Z, params=None)
         fcov = self.get_post_fcov(X, Z, params=None, diag=False)
-        return np.random.multivariate_normal(fmean, fcov * alpha**2, N)
+        return np.random.multivariate_normal(fmean, fcov * alpha ** 2, N)
 
     def predict_sampling(self, X, Z, params=None, N=1):
         """
@@ -411,3 +413,71 @@ def fit(self, X, t, config):
             params = bfgs.run(X, t)
 
         self.set_params(params)
+
+class GP_sfs(model):
+
+    def __init__(self, lik, mean, cov, xtrain=None, ytrain=None, inf="exact",config=None):
+        super().__init__(lik, mean, cov, xtrain, ytrain, inf)
+
+        self.config = config
+
+    def fit(self, X, t):
+        """
+        Fitting function (update parameters)
+
+        Parameters
+        ----------
+        X: numpy.ndarray
+            N x d dimensional matrix. Each row of X denotes the d-dimensional feature vector of search candidate.
+
+        t:  numpy.ndarray
+            N dimensional array.
+            The negative energy of each search candidate (value of the objective function to be optimized).
+        config: physbo.misc.set_config object
+
+        """
+        #config = set_config()
+        method = self.config.learning.method
+
+        if method == "adam":
+            adam = learning.adam(self, self.config)
+            params = adam.run(X, t)
+
+        if method in ("bfgs", "batch"):
+            bfgs = learning.batch(self, self.config)
+            params = bfgs.run(X, t)
+
+        self.set_params(params)
+
+        self.prepare(X, t, params=None)
+
+    def prepare(self, X, t, params=None):
+        return super().prepare(X, t, params)
+    
+    def get_post_fmean(self, X, Z, params=None):
+        return super().get_post_fmean(X, Z, params)
+
+    def predict(self, Z, params=None):
+        """
+        Calculating posterior mean of model (function)
+
+        Parameters
+        ==========
+        X: numpy.ndarray
+            inputs
+        Z: numpy.ndarray
+            feature maps
+        params: numpy.ndarray
+            Parameters
+        See also
+        ========
+        physbo.gp.inf.exact.get_post_fmean
+        """
+
+        if params is None:
+            params = np.copy(self.params)
+
+        if self.inf == "exact":
+            post_fmu = inf.exact.get_post_fmean(self, Z, Z, params)
+
+        return post_fmu
\ No newline at end of file

From 56c581ee9207fe31099263a133d2de60f2e1b796 Mon Sep 17 00:00:00 2001
From: Li Hengyu <66453357+HengyuLi-Ozaki-lab@users.noreply.github.com>
Date: Mon, 9 Oct 2023 22:22:26 +0900
Subject: [PATCH 2/5] Add new function for cross validation

get_params is created for clone when using cross validation.
---
 physbo/gp/core/model.py | 29 +++++++++++++++++++----------
 1 file changed, 19 insertions(+), 10 deletions(-)

diff --git a/physbo/gp/core/model.py b/physbo/gp/core/model.py
index 64b692ef..8de2a255 100644
--- a/physbo/gp/core/model.py
+++ b/physbo/gp/core/model.py
@@ -5,9 +5,10 @@
 from physbo.gp.core import learning
 from physbo.gp.core.prior import prior
 from physbo.misc import set_config
+from mlxtend.feature_selection import SequentialFeatureSelector as SFS
 
 class model:
-    def __init__(self, lik, mean, cov, xtrain=None, ytrain=None, inf="exact"):
+    def __init__(self, lik, mean, cov, inf="exact"):
         """
 
         Parameters
@@ -20,8 +21,6 @@ def __init__(self, lik, mean, cov, xtrain=None, ytrain=None, inf="exact"):
         self.lik = lik
         self.prior = prior(mean=mean, cov=cov)
         self.inf = inf
-        self.xtrain = xtrain
-        self.ytrain = ytrain
         self.num_params = self.lik.num_params + self.prior.num_params
         self.params = self.cat_params(self.lik.params, self.prior.params)
         self.stats = ()
@@ -414,13 +413,16 @@ def fit(self, X, t, config):
 
         self.set_params(params)
 
-class GP_sfs(model):
+class sfs(model):
 
-    def __init__(self, lik, mean, cov, xtrain=None, ytrain=None, inf="exact",config=None):
-        super().__init__(lik, mean, cov, xtrain, ytrain, inf)
+    def __init__(self, lik, mean, cov, inf="exact",config=None):
+        super().__init__(lik, mean, cov, inf)
 
         self.config = config
 
+    def prepare(self, X, t, params=None):
+        return super().prepare(X, t, params)
+
     def fit(self, X, t):
         """
         Fitting function (update parameters)
@@ -451,8 +453,7 @@ def fit(self, X, t):
 
         self.prepare(X, t, params=None)
 
-    def prepare(self, X, t, params=None):
-        return super().prepare(X, t, params)
+        self.xtrain = X
     
     def get_post_fmean(self, X, Z, params=None):
         return super().get_post_fmean(X, Z, params)
@@ -478,6 +479,14 @@ def predict(self, Z, params=None):
             params = np.copy(self.params)
 
         if self.inf == "exact":
-            post_fmu = inf.exact.get_post_fmean(self, Z, Z, params)
+            post_fmu = inf.exact.get_post_fmean(self, self.xtrain, Z, params)
+
+        return post_fmu
+    
+    def get_params(self,deep=True):
+
+        mean = self.prior.mean
+        cov = self.prior.cov
+        config = self.config
 
-        return post_fmu
\ No newline at end of file
+        return {"lik":self.lik,"mean":mean,"cov":cov,"config":config}
\ No newline at end of file

From e0b9042ad326ebcf06d3cd1c643b4b05e27c18c3 Mon Sep 17 00:00:00 2001
From: Li Hengyu <66453357+HengyuLi-Ozaki-lab@users.noreply.github.com>
Date: Mon, 9 Oct 2023 22:28:45 +0900
Subject: [PATCH 3/5] Update __init__.py

---
 physbo/gp/__init__.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/physbo/gp/__init__.py b/physbo/gp/__init__.py
index ee1bd702..50ca44f1 100644
--- a/physbo/gp/__init__.py
+++ b/physbo/gp/__init__.py
@@ -4,5 +4,6 @@
 
 from .core import prior
 from .core import model
+from .core import sfs
 from .core import learning
 from .predictor import predictor

From 144a26635b39437917c31e4eb232ea114553fc28 Mon Sep 17 00:00:00 2001
From: Li Hengyu <66453357+HengyuLi-Ozaki-lab@users.noreply.github.com>
Date: Mon, 9 Oct 2023 22:29:44 +0900
Subject: [PATCH 4/5] Update __init__.py

---
 physbo/gp/core/__init__.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/physbo/gp/core/__init__.py b/physbo/gp/core/__init__.py
index d2e597be..a00001bc 100644
--- a/physbo/gp/core/__init__.py
+++ b/physbo/gp/core/__init__.py
@@ -1,4 +1,5 @@
 from .prior import prior
 from .model import model
+from .model import sfs
 
 # from predictor import predictor

From 8dbf2b41ec0d77b5485721bc0077d6b8d9a11ca8 Mon Sep 17 00:00:00 2001
From: Li Hengyu <66453357+HengyuLi-Ozaki-lab@users.noreply.github.com>
Date: Mon, 9 Oct 2023 22:33:16 +0900
Subject: [PATCH 5/5] Test code for SFFS

Sequential forward floating selection combined with PHYSBO, here is a test code. The input parameter for SequentialFeatureSelector in mlxtend library can be find in:

https://rasbt.github.io/mlxtend/user_guide/feature_selection/SequentialFeatureSelector/
---
 examples/SFFS.ipynb | 1010 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 1010 insertions(+)
 create mode 100644 examples/SFFS.ipynb

diff --git a/examples/SFFS.ipynb b/examples/SFFS.ipynb
new file mode 100644
index 00000000..d225c81a
--- /dev/null
+++ b/examples/SFFS.ipynb
@@ -0,0 +1,1010 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import physbo\n",
+    "\n",
+    "import numpy as np\n",
+    "\n",
+    "from mlxtend.feature_selection import SequentialFeatureSelector as SFS\n",
+    "\n",
+    "# define target function\n",
+    "\n",
+    "def target(x1,x2,x3,x4,x5):\n",
+    "\n",
+    "    y = (100-x5)**2 + (2-x3)**2 + (10-x1)**2 + (1-x2)**2 + x4 + (1000-x1*x2) + np.random.normal(0,1)\n",
+    "\n",
+    "    return y\n",
+    "\n",
+    "para_num=2000\n",
+    "\n",
+    "coefficient1 = np.random.uniform(-10,10,para_num)\n",
+    "coefficient2 = np.random.uniform(-10,10,para_num)\n",
+    "coefficient3 = np.random.uniform(-10,10,para_num)\n",
+    "coefficient4 = np.random.uniform(-10,10,para_num)\n",
+    "coefficient5 = np.random.uniform(-10,10,para_num)\n",
+    "coefficient = np.array([coefficient1,coefficient2,coefficient3,coefficient4,coefficient5]).T # coefficient = features X = [[coefficient1,coefficient3],[coefficient3,coefficient1]]\n",
+    "\n",
+    "y = np.array([target(coefficient[i][0],coefficient[i][1],coefficient[i][2],coefficient[i][3],coefficient[i][4]) for i in range(para_num)])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "cov = physbo.gp.cov.gauss(coefficient.shape[1],ard = False)\n",
+    "\n",
+    "mean = physbo.gp.mean.const()\n",
+    "\n",
+    "lik = physbo.gp.lik.gauss()\n",
+    "\n",
+    "config = physbo.misc.set_config()\n",
+    "\n",
+    "gp = physbo.gp.sfs(lik=lik,mean=mean,cov=cov,config=config)\n",
+    "\n",
+    "test = physbo.gp.model(lik=lik,mean=mean,cov=cov)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sfs = SFS(estimator=gp,\n",
+    "          k_features=3,\n",
+    "          forward=True,\n",
+    "          floating=True,\n",
+    "          verbose=2,\n",
+    "          scoring='r2', \n",
+    "          cv=3,clone_estimator=False,n_jobs=-1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8394.692107640107\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 24961.10534143752\n",
+      "0 -th epoch marginal likelihood 25027.237945442175\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 24988.522313673915\n",
+      "0 -th epoch marginal likelihood 25058.811482450634\n",
+      "50 -th epoch marginal likelihood 8393.252454868598\n",
+      "50 -th epoch marginal likelihood 13615.070944271509\n",
+      "50 -th epoch marginal likelihood 13656.90991534373\n",
+      "50 -th epoch marginal likelihood 13641.503186631024\n",
+      "50 -th epoch marginal likelihood 13645.13132110246\n",
+      "100 -th epoch marginal likelihood 8392.291570455853\n",
+      "100 -th epoch marginal likelihood 11882.21265905766\n",
+      "100 -th epoch marginal likelihood 11891.191805530565\n",
+      "100 -th epoch marginal likelihood 11888.694027883696\n",
+      "100 -th epoch marginal likelihood 11889.644276096426\n",
+      "150 -th epoch marginal likelihood 8391.943793337241\n",
+      "150 -th epoch marginal likelihood 11438.129253146151\n",
+      "150 -th epoch marginal likelihood 11439.647584499286\n",
+      "150 -th epoch marginal likelihood 11440.231933927062\n",
+      "200 -th epoch marginal likelihood 8391.794064563468\n",
+      "150 -th epoch marginal likelihood 11440.112038417898\n",
+      "200 -th epoch marginal likelihood 11322.76624936791\n",
+      "200 -th epoch marginal likelihood 11325.29445371271\n",
+      "200 -th epoch marginal likelihood 11325.79022517724\n",
+      "250 -th epoch marginal likelihood 8391.771224361262\n",
+      "200 -th epoch marginal likelihood 11324.66858603153\n",
+      "250 -th epoch marginal likelihood 11300.884864349422\n",
+      "250 -th epoch marginal likelihood 11303.242319680172\n",
+      "250 -th epoch marginal likelihood 11303.847726637523\n",
+      "300 -th epoch marginal likelihood 8391.800527908683\n",
+      "250 -th epoch marginal likelihood 11302.101996308704\n",
+      "300 -th epoch marginal likelihood 11298.224933373665\n",
+      "300 -th epoch marginal likelihood 11300.401889133342\n",
+      "300 -th epoch marginal likelihood 11300.978415220317\n",
+      "350 -th epoch marginal likelihood 8391.772475113994\n",
+      "350 -th epoch marginal likelihood 11297.66712142268\n",
+      "300 -th epoch marginal likelihood 11299.22497720701\n",
+      "350 -th epoch marginal likelihood 11299.829019368835\n",
+      "350 -th epoch marginal likelihood 11300.27247702757\n",
+      "400 -th epoch marginal likelihood 8391.766783284382\n",
+      "400 -th epoch marginal likelihood 11297.43713817204\n",
+      "400 -th epoch marginal likelihood 11299.466772435286\n",
+      "350 -th epoch marginal likelihood 11298.646190715814\n",
+      "400 -th epoch marginal likelihood 11299.91841895285\n",
+      "450 -th epoch marginal likelihood 8391.779322204538\n",
+      "450 -th epoch marginal likelihood 11297.261613554807\n",
+      "450 -th epoch marginal likelihood 11299.133411050621\n",
+      "450 -th epoch marginal likelihood 11299.576252279945\n",
+      "400 -th epoch marginal likelihood 11298.368044657163\n",
+      "500 -th epoch marginal likelihood 8391.785505818381\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "500 -th epoch marginal likelihood 11297.144247816968\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 11298.976585080529\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "500 -th epoch marginal likelihood 11299.322207608308\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "450 -th epoch marginal likelihood 11298.1220460415\n",
+      "500 -th epoch marginal likelihood 11298.005702755341\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8437.131715888852\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25016.540404537067\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 24943.318098573644\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25073.018163957233\n",
+      "50 -th epoch marginal likelihood 8435.917491503891\n",
+      "50 -th epoch marginal likelihood 13630.446673030194\n",
+      "50 -th epoch marginal likelihood 13619.672955292608\n",
+      "50 -th epoch marginal likelihood 13638.295199293187\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25005.045588399225\n",
+      "100 -th epoch marginal likelihood 8435.118531586477\n",
+      "100 -th epoch marginal likelihood 11864.177128193796\n",
+      "100 -th epoch marginal likelihood 11861.671812960989\n",
+      "100 -th epoch marginal likelihood 11871.692812143678\n",
+      "50 -th epoch marginal likelihood 13613.372996947257\n",
+      "150 -th epoch marginal likelihood 8434.828055010901\n",
+      "150 -th epoch marginal likelihood 11410.586333116902\n",
+      "150 -th epoch marginal likelihood 11408.408663689872\n",
+      "150 -th epoch marginal likelihood 11413.273303823658\n",
+      "100 -th epoch marginal likelihood 11857.854507662345\n",
+      "200 -th epoch marginal likelihood 8434.69433935276\n",
+      "200 -th epoch marginal likelihood 11293.553614191824\n",
+      "200 -th epoch marginal likelihood 11292.307068175578\n",
+      "200 -th epoch marginal likelihood 11294.863949801447\n",
+      "150 -th epoch marginal likelihood 11405.124432818162\n",
+      "250 -th epoch marginal likelihood 8434.660576522681\n",
+      "250 -th epoch marginal likelihood 11270.965787251664\n",
+      "250 -th epoch marginal likelihood 11269.841529089403\n",
+      "250 -th epoch marginal likelihood 11271.544954824321\n",
+      "200 -th epoch marginal likelihood 11290.434923495066\n",
+      "300 -th epoch marginal likelihood 8434.660217124616\n",
+      "300 -th epoch marginal likelihood 11267.92454008275\n",
+      "300 -th epoch marginal likelihood 11266.916839272215\n",
+      "300 -th epoch marginal likelihood 11268.414177626493\n",
+      "250 -th epoch marginal likelihood 11268.679021146498\n",
+      "350 -th epoch marginal likelihood 11267.23699453149\n",
+      "350 -th epoch marginal likelihood 8434.673495603723\n",
+      "350 -th epoch marginal likelihood 11266.341709641243\n",
+      "350 -th epoch marginal likelihood 11267.656361725609\n",
+      "300 -th epoch marginal likelihood 11265.9453044816\n",
+      "400 -th epoch marginal likelihood 8434.627581516836\n",
+      "400 -th epoch marginal likelihood 11266.833038624958\n",
+      "400 -th epoch marginal likelihood 11265.93125294241\n",
+      "400 -th epoch marginal likelihood 11267.168699955106\n",
+      "350 -th epoch marginal likelihood 11265.476711890784\n",
+      "450 -th epoch marginal likelihood 8434.648234149154\n",
+      "450 -th epoch marginal likelihood 11266.520644059787\n",
+      "450 -th epoch marginal likelihood 11265.710905854541\n",
+      "450 -th epoch marginal likelihood 11266.778574564125\n",
+      "400 -th epoch marginal likelihood 11265.169921183297\n",
+      "500 -th epoch marginal likelihood 8434.666100385348\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 11266.331639298953\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "500 -th epoch marginal likelihood 11265.583531605695\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 11266.50722506471\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "450 -th epoch marginal likelihood 11264.989918808309\n",
+      "500 -th epoch marginal likelihood 11264.890974271668\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8426.917442722235\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25055.030913118226\n",
+      "0 -th epoch marginal likelihood 24945.672301020117\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25069.40362965293\n",
+      "50 -th epoch marginal likelihood 8425.546079727721\n",
+      "50 -th epoch marginal likelihood 13657.424110022224\n",
+      "50 -th epoch marginal likelihood 13626.032879292772\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25049.77648975408\n",
+      "50 -th epoch marginal likelihood 13664.89911623701\n",
+      "100 -th epoch marginal likelihood 8424.880338206025\n",
+      "100 -th epoch marginal likelihood 11902.62730404944\n",
+      "100 -th epoch marginal likelihood 11895.750624808115\n",
+      "100 -th epoch marginal likelihood 11905.516755759549\n",
+      "50 -th epoch marginal likelihood 13657.382221036096\n",
+      "150 -th epoch marginal likelihood 8424.583063882486\n",
+      "150 -th epoch marginal likelihood 11452.52209895657\n",
+      "150 -th epoch marginal likelihood 11448.45254163338\n",
+      "150 -th epoch marginal likelihood 11454.336366838746\n",
+      "100 -th epoch marginal likelihood 11903.247090762821\n",
+      "200 -th epoch marginal likelihood 11337.984767393193\n",
+      "200 -th epoch marginal likelihood 8424.464509187645\n",
+      "200 -th epoch marginal likelihood 11333.275511860928\n",
+      "200 -th epoch marginal likelihood 11338.420775516182\n",
+      "150 -th epoch marginal likelihood 11452.939713954627\n",
+      "250 -th epoch marginal likelihood 11315.576616625054\n",
+      "250 -th epoch marginal likelihood 8424.415266105316\n",
+      "250 -th epoch marginal likelihood 11311.398562302604\n",
+      "250 -th epoch marginal likelihood 11316.204969405153\n",
+      "200 -th epoch marginal likelihood 11337.677305119947\n",
+      "300 -th epoch marginal likelihood 11312.535005554357\n",
+      "300 -th epoch marginal likelihood 8424.426948175791\n",
+      "300 -th epoch marginal likelihood 11308.815512707226\n",
+      "300 -th epoch marginal likelihood 11313.111728720203\n",
+      "250 -th epoch marginal likelihood 11315.146957037115\n",
+      "350 -th epoch marginal likelihood 11311.93706358209\n",
+      "350 -th epoch marginal likelihood 8424.4084640009\n",
+      "350 -th epoch marginal likelihood 11308.365236463476\n",
+      "300 -th epoch marginal likelihood 11312.423540961854\n",
+      "350 -th epoch marginal likelihood 11312.442852595052\n",
+      "400 -th epoch marginal likelihood 11311.5168885129\n",
+      "400 -th epoch marginal likelihood 8424.42263191808\n",
+      "400 -th epoch marginal likelihood 11308.205124504831\n",
+      "400 -th epoch marginal likelihood 11312.038173816965\n",
+      "350 -th epoch marginal likelihood 11311.825860380113\n",
+      "450 -th epoch marginal likelihood 11311.275951176323\n",
+      "450 -th epoch marginal likelihood 11308.139396131613\n",
+      "450 -th epoch marginal likelihood 8424.457138682741\n",
+      "450 -th epoch marginal likelihood 11311.626448257965\n",
+      "400 -th epoch marginal likelihood 11311.528324244358\n",
+      "500 -th epoch marginal likelihood 11311.087798336908\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 8424.42713327391\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 11308.100460086758\n",
+      "Done\n",
+      "\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   32.4s remaining:   48.6s\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "500 -th epoch marginal likelihood 11311.38520824873\n",
+      "Done\n",
+      "\n",
+      "450 -th epoch marginal likelihood 11311.267549038928\n",
+      "500 -th epoch marginal likelihood 11311.071330999132\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   32.8s finished\n",
+      "\n",
+      "[2023-10-09 22:11:10] Features: 1/3 -- score: 0.9868212005393535[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8348.661884201516\n",
+      "0 -th epoch marginal likelihood 8406.156516774883\n",
+      "50 -th epoch marginal likelihood 8343.172238132636\n",
+      "50 -th epoch marginal likelihood 8400.466800380731\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 7366.98853938562\n",
+      "0 -th epoch marginal likelihood 8370.15611847102\n",
+      "100 -th epoch marginal likelihood 8341.721021871012\n",
+      "100 -th epoch marginal likelihood 8398.870580527268\n",
+      "50 -th epoch marginal likelihood 8364.842902199669\n",
+      "50 -th epoch marginal likelihood 7358.517947131348\n",
+      "150 -th epoch marginal likelihood 8341.231750204237\n",
+      "150 -th epoch marginal likelihood 8398.328133055265\n",
+      "100 -th epoch marginal likelihood 7356.620976941095\n",
+      "100 -th epoch marginal likelihood 8363.371811862879\n",
+      "200 -th epoch marginal likelihood 8341.053793463583\n",
+      "200 -th epoch marginal likelihood 8398.11729552888\n",
+      "150 -th epoch marginal likelihood 7356.05737114786\n",
+      "150 -th epoch marginal likelihood 8362.868906476739\n",
+      "250 -th epoch marginal likelihood 8340.98283159921\n",
+      "250 -th epoch marginal likelihood 8398.027422725974\n",
+      "200 -th epoch marginal likelihood 7355.704667154792\n",
+      "200 -th epoch marginal likelihood 8362.757006347629\n",
+      "300 -th epoch marginal likelihood 8340.988043231036\n",
+      "300 -th epoch marginal likelihood 8397.978815952676\n",
+      "250 -th epoch marginal likelihood 7355.6064865684575\n",
+      "250 -th epoch marginal likelihood 8362.66835436513\n",
+      "350 -th epoch marginal likelihood 8340.959273121345\n",
+      "350 -th epoch marginal likelihood 8397.979738254757\n",
+      "300 -th epoch marginal likelihood 7355.745602799287\n",
+      "300 -th epoch marginal likelihood 8362.654043939474\n",
+      "400 -th epoch marginal likelihood 8340.964442447279\n",
+      "400 -th epoch marginal likelihood 8397.935617859379\n",
+      "350 -th epoch marginal likelihood 7355.5908640773105\n",
+      "350 -th epoch marginal likelihood 8362.589374987134\n",
+      "450 -th epoch marginal likelihood 8340.941011928955\n",
+      "450 -th epoch marginal likelihood 8397.922297035038\n",
+      "400 -th epoch marginal likelihood 7355.613606565945\n",
+      "400 -th epoch marginal likelihood 8362.593986901027\n",
+      "500 -th epoch marginal likelihood 8340.93336736056\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 8397.940346136087\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "450 -th epoch marginal likelihood 7355.637140339025\n",
+      "450 -th epoch marginal likelihood 8362.575630404288\n",
+      "500 -th epoch marginal likelihood 7355.636810652959\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 8362.56983824255\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8448.799498703494\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8384.64530148956\n",
+      "50 -th epoch marginal likelihood 8443.323547470913\n",
+      "50 -th epoch marginal likelihood 8379.449153339217\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8410.899084508474\n",
+      "0 -th epoch marginal likelihood 7372.625087352935\n",
+      "100 -th epoch marginal likelihood 8441.804850238816\n",
+      "100 -th epoch marginal likelihood 8378.054781075958\n",
+      "50 -th epoch marginal likelihood 8405.761183377332\n",
+      "50 -th epoch marginal likelihood 7364.866891824461\n",
+      "150 -th epoch marginal likelihood 8441.365082381024\n",
+      "150 -th epoch marginal likelihood 8377.62888318411\n",
+      "100 -th epoch marginal likelihood 8404.28124725758\n",
+      "100 -th epoch marginal likelihood 7363.194110903938\n",
+      "200 -th epoch marginal likelihood 8441.195836849569\n",
+      "200 -th epoch marginal likelihood 8377.494438279977\n",
+      "150 -th epoch marginal likelihood 8403.80385289546\n",
+      "150 -th epoch marginal likelihood 7362.591366863666\n",
+      "250 -th epoch marginal likelihood 8441.11929407143\n",
+      "250 -th epoch marginal likelihood 8377.415850985173\n",
+      "200 -th epoch marginal likelihood 8403.66286455045\n",
+      "200 -th epoch marginal likelihood 7362.462823929736\n",
+      "300 -th epoch marginal likelihood 8441.140771998294\n",
+      "300 -th epoch marginal likelihood 8377.429238495792\n",
+      "250 -th epoch marginal likelihood 8403.603933290527\n",
+      "250 -th epoch marginal likelihood 7362.438292943982\n",
+      "350 -th epoch marginal likelihood 8441.106446557207\n",
+      "350 -th epoch marginal likelihood 8377.411044560164\n",
+      "300 -th epoch marginal likelihood 7362.428802144553\n",
+      "300 -th epoch marginal likelihood 8403.635201940566\n",
+      "400 -th epoch marginal likelihood 8441.178459103154\n",
+      "400 -th epoch marginal likelihood 8377.445780564696\n",
+      "350 -th epoch marginal likelihood 7362.425638400319\n",
+      "350 -th epoch marginal likelihood 8403.543806552287\n",
+      "450 -th epoch marginal likelihood 8441.09870629876\n",
+      "450 -th epoch marginal likelihood 8377.403338715741\n",
+      "400 -th epoch marginal likelihood 8403.561989566668\n",
+      "400 -th epoch marginal likelihood 7362.49334292305\n",
+      "500 -th epoch marginal likelihood 8441.111928822826\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "500 -th epoch marginal likelihood 8377.393454337565\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "450 -th epoch marginal likelihood 8403.522918487286\n",
+      "450 -th epoch marginal likelihood 7362.325613212061\n",
+      "500 -th epoch marginal likelihood 8403.52780370436\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 7362.392479889024\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8438.64709433931\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8369.226746477078\n",
+      "50 -th epoch marginal likelihood 8432.650151554559\n",
+      "50 -th epoch marginal likelihood 8363.96339380385\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 7416.292266847133\n",
+      "0 -th epoch marginal likelihood 8408.717425059125\n",
+      "100 -th epoch marginal likelihood 8430.733744732877\n",
+      "100 -th epoch marginal likelihood 8362.581191517354\n",
+      "50 -th epoch marginal likelihood 7409.126474914163\n",
+      "150 -th epoch marginal likelihood 8430.147362588723\n",
+      "50 -th epoch marginal likelihood 8403.244621571883\n",
+      "150 -th epoch marginal likelihood 8362.164145382032\n",
+      "100 -th epoch marginal likelihood 7407.4852564482035\n",
+      "200 -th epoch marginal likelihood 8429.886804422102\n",
+      "100 -th epoch marginal likelihood 8401.59009400104\n",
+      "200 -th epoch marginal likelihood 8362.006436794016\n",
+      "150 -th epoch marginal likelihood 7406.971494733303\n",
+      "250 -th epoch marginal likelihood 8429.747527636157\n",
+      "150 -th epoch marginal likelihood 8401.011990622485\n",
+      "250 -th epoch marginal likelihood 8361.98702411757\n",
+      "200 -th epoch marginal likelihood 7406.769707951678\n",
+      "300 -th epoch marginal likelihood 8429.688112568621\n",
+      "200 -th epoch marginal likelihood 8400.793678112668\n",
+      "300 -th epoch marginal likelihood 8361.908527972973\n",
+      "250 -th epoch marginal likelihood 7406.679082044225\n",
+      "350 -th epoch marginal likelihood 8429.709924524612\n",
+      "250 -th epoch marginal likelihood 8400.707051416968\n",
+      "350 -th epoch marginal likelihood 8361.917387571768\n",
+      "300 -th epoch marginal likelihood 7406.653889771927\n",
+      "400 -th epoch marginal likelihood 8429.724953474339\n",
+      "300 -th epoch marginal likelihood 8400.662601247699\n",
+      "400 -th epoch marginal likelihood 8361.965621119074\n",
+      "350 -th epoch marginal likelihood 7406.73671764464\n",
+      "450 -th epoch marginal likelihood 8429.712014970715\n",
+      "350 -th epoch marginal likelihood 8400.656766295002\n",
+      "450 -th epoch marginal likelihood 8361.940628673123\n",
+      "400 -th epoch marginal likelihood 7406.67590507235\n",
+      "500 -th epoch marginal likelihood 8429.676077994942\n",
+      "Done\n",
+      "\n",
+      "400 -th epoch marginal likelihood 8400.650924452952\n",
+      "500 -th epoch marginal likelihood 8361.99410020812\n",
+      "Done\n",
+      "\n",
+      "450 -th epoch marginal likelihood 7406.637415256406\n",
+      "450 -th epoch marginal likelihood 8400.65902380914\n",
+      "500 -th epoch marginal likelihood 7406.64668390076\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 8400.661119354912\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done   4 out of   4 | elapsed:   30.1s finished\n",
+      "\n",
+      "[2023-10-09 22:11:41] Features: 2/3 -- score: 0.9972916731489968[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 7380.252052642909\n",
+      "0 -th epoch marginal likelihood 6853.388069440912\n",
+      "0 -th epoch marginal likelihood 7072.274491559429\n",
+      "50 -th epoch marginal likelihood 7366.320818436539\n",
+      "50 -th epoch marginal likelihood 6768.635202201106\n",
+      "50 -th epoch marginal likelihood 7042.191714856888\n",
+      "100 -th epoch marginal likelihood 7363.401371837834\n",
+      "100 -th epoch marginal likelihood 6765.818993157858\n",
+      "100 -th epoch marginal likelihood 7039.314956732355\n",
+      "150 -th epoch marginal likelihood 7362.5339645011145\n",
+      "150 -th epoch marginal likelihood 6765.093795292746\n",
+      "150 -th epoch marginal likelihood 7038.508187698322\n",
+      "200 -th epoch marginal likelihood 7362.175211895561\n",
+      "200 -th epoch marginal likelihood 6764.801199017752\n",
+      "200 -th epoch marginal likelihood 7038.2646571703435\n",
+      "250 -th epoch marginal likelihood 7362.195496797939\n",
+      "250 -th epoch marginal likelihood 6764.79219024564\n",
+      "250 -th epoch marginal likelihood 7038.093235296676\n",
+      "300 -th epoch marginal likelihood 7362.031643175413\n",
+      "300 -th epoch marginal likelihood 6764.787875118791\n",
+      "300 -th epoch marginal likelihood 7038.043072232177\n",
+      "350 -th epoch marginal likelihood 7362.018293257057\n",
+      "350 -th epoch marginal likelihood 6764.659697138307\n",
+      "350 -th epoch marginal likelihood 7037.946251147738\n",
+      "400 -th epoch marginal likelihood 7361.984553140729\n",
+      "400 -th epoch marginal likelihood 6764.716380702993\n",
+      "400 -th epoch marginal likelihood 7038.055586421422\n",
+      "450 -th epoch marginal likelihood 7361.973581031396\n",
+      "450 -th epoch marginal likelihood 6764.782999717808\n",
+      "450 -th epoch marginal likelihood 7038.061743475627\n",
+      "500 -th epoch marginal likelihood 7362.004635776717\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 6764.823031644267\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 7038.043200140586\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 7383.934785966861\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 6853.537518160264\n",
+      "0 -th epoch marginal likelihood 7057.200729919182\n",
+      "50 -th epoch marginal likelihood 7372.9222462359485\n",
+      "50 -th epoch marginal likelihood 6781.992509001286\n",
+      "50 -th epoch marginal likelihood 7030.031780114455\n",
+      "100 -th epoch marginal likelihood 7370.619133823647\n",
+      "100 -th epoch marginal likelihood 6779.576662201424\n",
+      "100 -th epoch marginal likelihood 7027.830999356948\n",
+      "150 -th epoch marginal likelihood 7370.105054755204\n",
+      "150 -th epoch marginal likelihood 6779.001180218182\n",
+      "150 -th epoch marginal likelihood 7026.5028676593565\n",
+      "200 -th epoch marginal likelihood 7369.967150907525\n",
+      "200 -th epoch marginal likelihood 6778.746276112818\n",
+      "200 -th epoch marginal likelihood 7026.504799517027\n",
+      "250 -th epoch marginal likelihood 7369.950236525045\n",
+      "250 -th epoch marginal likelihood 6778.788134673748\n",
+      "250 -th epoch marginal likelihood 7026.0146679666595\n",
+      "300 -th epoch marginal likelihood 7369.913378081135\n",
+      "300 -th epoch marginal likelihood 6778.598192836724\n",
+      "300 -th epoch marginal likelihood 7025.760779436226\n",
+      "350 -th epoch marginal likelihood 7369.89229989129\n",
+      "350 -th epoch marginal likelihood 6778.655385232025\n",
+      "350 -th epoch marginal likelihood 7026.231836195792\n",
+      "400 -th epoch marginal likelihood 7369.880223932143\n",
+      "400 -th epoch marginal likelihood 6778.750184089111\n",
+      "400 -th epoch marginal likelihood 7026.206421882574\n",
+      "450 -th epoch marginal likelihood 7369.8818994944495\n",
+      "450 -th epoch marginal likelihood 6778.600793212785\n",
+      "450 -th epoch marginal likelihood 7026.169073314849\n",
+      "500 -th epoch marginal likelihood 7369.879932594257\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "500 -th epoch marginal likelihood 6778.650064423405\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 7025.985606009753\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 7425.236133784977\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 6882.4801596008565\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 7116.144131978775\n",
+      "50 -th epoch marginal likelihood 7412.079020766404\n",
+      "50 -th epoch marginal likelihood 6802.732401001239\n",
+      "50 -th epoch marginal likelihood 7088.515315448793\n",
+      "100 -th epoch marginal likelihood 7408.999897897296\n",
+      "100 -th epoch marginal likelihood 6799.809368505202\n",
+      "100 -th epoch marginal likelihood 7085.110118427161\n",
+      "150 -th epoch marginal likelihood 7408.140727804241\n",
+      "150 -th epoch marginal likelihood 6799.07291074185\n",
+      "150 -th epoch marginal likelihood 7083.84619195766\n",
+      "200 -th epoch marginal likelihood 7407.856791237282\n",
+      "200 -th epoch marginal likelihood 6798.790111335462\n",
+      "200 -th epoch marginal likelihood 7083.592439544232\n",
+      "250 -th epoch marginal likelihood 7407.7265079654535\n",
+      "250 -th epoch marginal likelihood 6798.703666439378\n",
+      "250 -th epoch marginal likelihood 7083.615407897124\n",
+      "300 -th epoch marginal likelihood 7407.679377148817\n",
+      "300 -th epoch marginal likelihood 6798.682939252364\n",
+      "300 -th epoch marginal likelihood 7083.43114420494\n",
+      "350 -th epoch marginal likelihood 7407.698426318327\n",
+      "350 -th epoch marginal likelihood 6798.696704051623\n",
+      "350 -th epoch marginal likelihood 7083.37901855841\n",
+      "400 -th epoch marginal likelihood 7407.757573212317\n",
+      "400 -th epoch marginal likelihood 6798.8052843951555\n",
+      "400 -th epoch marginal likelihood 7083.456279076076\n",
+      "450 -th epoch marginal likelihood 7407.679463501584\n",
+      "450 -th epoch marginal likelihood 6798.702033160487\n",
+      "450 -th epoch marginal likelihood 7083.543513835316\n",
+      "500 -th epoch marginal likelihood 7407.631113131754\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 6798.839509680971\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 7083.436575167636\n",
+      "Done\n",
+      "\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done   3 out of   3 | elapsed:   27.1s finished\n",
+      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25417.25952973674\n",
+      "0 -th epoch marginal likelihood 8370.150691315843\n",
+      "50 -th epoch marginal likelihood 14568.545907200218\n",
+      "50 -th epoch marginal likelihood 8364.753935882698\n",
+      "100 -th epoch marginal likelihood 12862.33467989464\n",
+      "100 -th epoch marginal likelihood 8363.30442487214\n",
+      "150 -th epoch marginal likelihood 11986.867433425134\n",
+      "150 -th epoch marginal likelihood 8362.85102673119\n",
+      "200 -th epoch marginal likelihood 11558.290423093193\n",
+      "200 -th epoch marginal likelihood 8362.692841194772\n",
+      "250 -th epoch marginal likelihood 11383.269424890985\n",
+      "250 -th epoch marginal likelihood 8362.61693373179\n",
+      "300 -th epoch marginal likelihood 11327.535662659871\n",
+      "300 -th epoch marginal likelihood 8362.60493075461\n",
+      "350 -th epoch marginal likelihood 11308.248221165366\n",
+      "350 -th epoch marginal likelihood 8362.554338698534\n",
+      "400 -th epoch marginal likelihood 11301.288097330098\n",
+      "400 -th epoch marginal likelihood 8362.569360967344\n",
+      "450 -th epoch marginal likelihood 11298.13683732899\n",
+      "450 -th epoch marginal likelihood 8362.588251109968\n",
+      "500 -th epoch marginal likelihood 11296.01073859547\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 8362.559431238897\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25421.881394094737\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8410.918451174932\n",
+      "50 -th epoch marginal likelihood 14339.941367061208\n",
+      "50 -th epoch marginal likelihood 8405.847980735656\n",
+      "100 -th epoch marginal likelihood 12625.850659742582\n",
+      "100 -th epoch marginal likelihood 8404.358429588547\n",
+      "150 -th epoch marginal likelihood 11890.892838746895\n",
+      "150 -th epoch marginal likelihood 8403.924370340072\n",
+      "200 -th epoch marginal likelihood 11522.778933305734\n",
+      "200 -th epoch marginal likelihood 8403.694224328197\n",
+      "250 -th epoch marginal likelihood 11365.346182291476\n",
+      "250 -th epoch marginal likelihood 8403.609327915476\n",
+      "300 -th epoch marginal likelihood 11306.706662766901\n",
+      "300 -th epoch marginal likelihood 8403.616597512182\n",
+      "350 -th epoch marginal likelihood 11286.464721719973\n",
+      "350 -th epoch marginal likelihood 8403.550724891156\n",
+      "400 -th epoch marginal likelihood 11278.90551489383\n",
+      "400 -th epoch marginal likelihood 8403.5438051382\n",
+      "450 -th epoch marginal likelihood 11274.414188707166\n",
+      "450 -th epoch marginal likelihood 8403.546835147039\n",
+      "500 -th epoch marginal likelihood 11271.236321882174\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "500 -th epoch marginal likelihood 8403.545799030224\n",
+      "Done\n",
+      "\n",
+      "Start the initial hyper parameter searching ...\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 25497.968812217412\n",
+      "Done\n",
+      "\n",
+      "Start the hyper parameter learning ...\n",
+      "0 -th epoch marginal likelihood 8408.705472676626\n",
+      "50 -th epoch marginal likelihood 14296.13346995813\n",
+      "50 -th epoch marginal likelihood 8403.239744736737\n",
+      "100 -th epoch marginal likelihood 12662.484300091477\n",
+      "100 -th epoch marginal likelihood 8401.626452874256\n",
+      "150 -th epoch marginal likelihood 11931.188047845548\n",
+      "150 -th epoch marginal likelihood 8401.033529280608\n",
+      "200 -th epoch marginal likelihood 11564.647303657104\n",
+      "200 -th epoch marginal likelihood 8400.841585609109\n",
+      "250 -th epoch marginal likelihood 11407.140678401727\n",
+      "250 -th epoch marginal likelihood 8400.740274067624\n",
+      "300 -th epoch marginal likelihood 11346.927779000993\n",
+      "300 -th epoch marginal likelihood 8400.726023845597\n",
+      "350 -th epoch marginal likelihood 11326.64090238014\n",
+      "350 -th epoch marginal likelihood 8400.690824674211\n",
+      "400 -th epoch marginal likelihood 11318.576422682858\n",
+      "400 -th epoch marginal likelihood 8400.753422347942\n",
+      "450 -th epoch marginal likelihood 11314.908946073198\n",
+      "450 -th epoch marginal likelihood 8400.706229365529\n",
+      "500 -th epoch marginal likelihood 11313.245342520775\n",
+      "Done\n",
+      "\n",
+      "500 -th epoch marginal likelihood 8400.67014871078\n",
+      "Done\n",
+      "\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "[Parallel(n_jobs=-1)]: Done   2 out of   2 | elapsed:   24.6s finished\n",
+      "\n",
+      "[2023-10-09 22:12:32] Features: 3/3 -- score: 0.9989209394666663"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<style>#sk-container-id-5 {color: black;}#sk-container-id-5 pre{padding: 0;}#sk-container-id-5 div.sk-toggleable {background-color: white;}#sk-container-id-5 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-5 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-5 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-5 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-5 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-5 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-5 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-5 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-5 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-5 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-5 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-5 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-5 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-5 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-5 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-5 div.sk-item {position: relative;z-index: 1;}#sk-container-id-5 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-5 div.sk-item::before, #sk-container-id-5 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-5 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-5 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-5 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-5 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-5 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-5 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-5 div.sk-label-container {text-align: center;}#sk-container-id-5 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-5 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-5\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SequentialFeatureSelector(clone_estimator=False, cv=3,\n",
+       "                          estimator=&lt;physbo.gp.core.model.GP_sfs object at 0x15b08c0a0&gt;,\n",
+       "                          floating=True, k_features=(3, 3), n_jobs=-1,\n",
+       "                          scoring=&#x27;r2&#x27;, verbose=2)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SequentialFeatureSelector</label><div class=\"sk-toggleable__content\"><pre>SequentialFeatureSelector(clone_estimator=False, cv=3,\n",
+       "                          estimator=&lt;physbo.gp.core.model.GP_sfs object at 0x15b08c0a0&gt;,\n",
+       "                          floating=True, k_features=(3, 3), n_jobs=-1,\n",
+       "                          scoring=&#x27;r2&#x27;, verbose=2)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: GP_sfs</label><div class=\"sk-toggleable__content\"><pre>&lt;physbo.gp.core.model.GP_sfs object at 0x15b08c0a0&gt;</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-15\" type=\"checkbox\" ><label for=\"sk-estimator-id-15\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GP_sfs</label><div class=\"sk-toggleable__content\"><pre>&lt;physbo.gp.core.model.GP_sfs object at 0x15b08c0a0&gt;</pre></div></div></div></div></div></div></div></div></div></div>"
+      ],
+      "text/plain": [
+       "SequentialFeatureSelector(clone_estimator=False, cv=3,\n",
+       "                          estimator=<physbo.gp.core.model.GP_sfs object at 0x15b08c0a0>,\n",
+       "                          floating=True, k_features=(3, 3), n_jobs=-1,\n",
+       "                          scoring='r2', verbose=2)"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sfs.fit(coefficient,y)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{1: {'feature_idx': (4,),\n",
+       "  'cv_scores': array([0.98557907, 0.98842159, 0.98646293]),\n",
+       "  'avg_score': 0.9868212005393535,\n",
+       "  'feature_names': ('4',)},\n",
+       " 2: {'feature_idx': (0, 4),\n",
+       "  'cv_scores': array([0.99707211, 0.99737709, 0.99742582]),\n",
+       "  'avg_score': 0.9972916731489968,\n",
+       "  'feature_names': ('0', '4')},\n",
+       " 3: {'feature_idx': (0, 1, 4),\n",
+       "  'cv_scores': array([0.99884154, 0.99898918, 0.9989321 ]),\n",
+       "  'avg_score': 0.9989209394666663,\n",
+       "  'feature_names': ('0', '1', '4')}}"
+      ]
+     },
+     "execution_count": 25,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sfs.subsets_ # CV score should converge to 1 # Test directly GP with 1 feature"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from mlxtend.plotting import plot_sequential_feature_selection as plot_sfs\n",
+    "\n",
+    "fig = plot_sfs(sfs.get_metric_dict(), kind='std_err')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{1: {'feature_idx': (4,),\n",
+       "  'cv_scores': array([0.98557907, 0.98842159, 0.98646293]),\n",
+       "  'avg_score': 0.9868212005393535,\n",
+       "  'feature_names': ('4',),\n",
+       "  'ci_bound': 0.0026729055301516714,\n",
+       "  'std_dev': 0.0011877841412142533,\n",
+       "  'std_err': 0.0008398902208384383},\n",
+       " 2: {'feature_idx': (0, 4),\n",
+       "  'cv_scores': array([0.99707211, 0.99737709, 0.99742582]),\n",
+       "  'avg_score': 0.9972916731489968,\n",
+       "  'feature_names': ('0', '4'),\n",
+       "  'ci_bound': 0.000352235687586893,\n",
+       "  'std_dev': 0.0001565262815935246,\n",
+       "  'std_err': 0.00011068079514869633},\n",
+       " 3: {'feature_idx': (0, 1, 4),\n",
+       "  'cv_scores': array([0.99884154, 0.99898918, 0.9989321 ]),\n",
+       "  'avg_score': 0.9989209394666663,\n",
+       "  'feature_names': ('0', '1', '4'),\n",
+       "  'ci_bound': 0.00013679366761352237,\n",
+       "  'std_dev': 6.0788287194218874e-05,\n",
+       "  'std_err': 4.298381009174754e-05}}"
+      ]
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sfs.get_metric_dict()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.6"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}