unhashable type: 'numpy.ndarray'を直したい！

前提・実現したいこと

TypeErrorを直したい！
また,解決法を教えてほしい.

発生している問題・エラーメッセージ

TypeError                                 Traceback (most recent call last)
<ipython-input-61-312f5a8ebc80> in <module>
     55     }
     56  # 上記のパラメータでモデルを学習する
---> 57 model = lgb.train(lgbm_params, lgb_train, valid_sets=lgb_eval)
     58 
     59     # テストデータを予測する

C:\ProgramData\Anaconda3\lib\site-packages\lightgbm\engine.py in train(params, train_set, num_boost_round, valid_sets, valid_names, fobj, feval, init_model, feature_name, categorical_feature, early_stopping_rounds, evals_result, verbose_eval, learning_rates, keep_training_booster, callbacks)
    160             if not isinstance(valid_data, Dataset):
    161                 raise TypeError("Traninig only accepts Dataset object")
--> 162             reduced_valid_sets.append(valid_data._update_params(params).set_reference(train_set))
    163             if valid_names is not None and len(valid_names) > i:
    164                 name_valid_sets.append(valid_names[i])

C:\ProgramData\Anaconda3\lib\site-packages\lightgbm\basic.py in set_reference(self, reference)
   1246             ._set_predictor(reference._predictor)
   1247         # we're done if self and reference share a common upstrem reference
-> 1248         if self.get_ref_chain().intersection(reference.get_ref_chain()):
   1249             return self
   1250         if self.data is not None:

C:\ProgramData\Anaconda3\lib\site-packages\lightgbm\basic.py in get_ref_chain(self, ref_limit)
   1487             if isinstance(head, Dataset):
   1488                 ref_chain.add(head)
-> 1489                 if (head.reference is not None) and (head.reference not in ref_chain):
   1490                     head = head.reference
   1491                 else:

TypeError: unhashable type: 'numpy.ndarray'

該当のソースコード

%matplotlib inline
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import lightgbm as lgb
from pandas import DataFrame, Series
from sklearn.datasets import make_blobs
from matplotlib.font_manager import FontProperties
fp = FontProperties(fname=r'c:\windows\Fonts\meiryo.ttc', size=14)
from sklearn.svm import SVC
from sklearn.metrics import r2_score
from sklearn.metrics import mean_squared_error
from sklearn.ensemble import RandomForestRegressor
import mglearn
data_1 = pd.read_csv("seiri.csv", header = None, encoding="UTF8")
pre_ent = np.array(data_1.iloc[0:6415,0])
pre_exi = np.array(data_1.iloc[0:6415,1])
flow_rate = np.array(data_1.iloc[0:6415,2])

pre_diff = pre_ent - pre_exi

Pre_Diff = pre_diff.reshape(1,-1)
Flow_Rate = flow_rate.reshape(1,-1)

Pre_Diff = pre_diff.reshape(1,-1)
Flow_Rate = flow_rate.reshape(1,-1)
X_train, X_test, Y_train, Y_test = train_test_split(Pre_Diff, Flow_Rate,train_size = 0.7)
x_train = np.linspace(0, 450, 100)
X_train = x_train.reshape(1,-1)
y_train = np.linspace(0, 4500, 50)
Y_train = x_train.reshape(1,-1)

z_train = np.cos(X_train)+(10**(-1))*(np.sin(-1)*(Y_train))-(X_train)/1.5--(Y_train)/15+550

lgb_train = lgb.Dataset(X_train, Y_train, z_train)




Z_test= np.linspace(550,0,100)
z_test = Z_test.reshape(1,-1)

lgb_eval = lgb.Dataset(X_test, Y_test, z_test)

 # LightGBM のハイパーパラメータ
lgbm_params = {
        # 回帰問題
        'objective': 'regression',
        
        'metric': 'r2_score',
    }
 # 上記のパラメータでモデルを学習する
model = lgb.train(lgbm_params, lgb_train, valid_sets=lgb_eval)

    # テストデータを予測する
z_pred = model.predict(X_test, Y_test, num_iteration=model.best_iteration)