用nni调参，不能正常运行

Question

问题遇到的现象和发生背景

用nni调参，不能正常运行，参考帖子：https://blog.csdn.net/qq_42103091/article/details/123287865?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522166538756916781432919967%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fall.%2522%257D&request_id=166538756916781432919967&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~first_rank_ecpm_v1~rank_v31_ecpm-11-123287865-null-null.142^v52^control,201^v3^add_ask&utm_term=nni%E5%AE%89%E8%A3%85&spm=1018.2226.3001.4187
在运行结果展示这一步，输入

nnictl create --config exp_config.yaml

然后去他给的链接去查看运行结果，结果都运行失败了，不知道哪里出了问题

用代码块功能插入代码，请勿粘贴截图

# -*- coding: utf-8 -*-
"""
Created on Mon Oct 10 20:41:44 2022

@author: fhm
"""


from tensorflow.keras import layers, optimizers, models
from tensorflow.keras.utils import to_categorical
from keras.datasets import mnist
from sklearn.metrics import accuracy_score
import nni


# 获取数据
def load_data():
    (train_data, train_labels), (test_data, test_labels) = mnist.load_data()  # 载入数据集
    # 数据集的归一化
    train_data = train_data.reshape((60000, 28, 28, 1))
    train_data = train_data.astype('float32') / 255
    test_data = test_data.reshape((10000, 28, 28, 1))
    test_data = test_data.astype('float32') / 255
    # 目标值的独热编码
    train_labels = to_categorical(train_labels)
    test_labels = to_categorical(test_labels)
    return train_data, train_labels, test_data, test_labels

def get_default_parameters():
    params = {'lr': 0.1}
    return params

# 搭建模型与模型编译
def get_model(params):
    model = models.Sequential()
    model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)))
    model.add(layers.MaxPooling2D((2, 2)))
    model.add(layers.Conv2D(64, (3, 3), activation='relu'))
    model.add(layers.MaxPooling2D((2, 2)))
    model.add(layers.Conv2D(64, (3, 3), activation='relu'))
    model.add(layers.Flatten())
    model.add(layers.Dense(64, activation='relu'))
    model.add(layers.Dense(10, activation='softmax'))
    lr = params.get('lr')
    model.compile(optimizer=optimizers.Adam(lr), loss='categorical_crossentropy', metrics=['accuracy'])
    return model
# 模型训练
def run(x_train, y_train, x_test, y_test, modelss):
    modelss.fit(x_train, y_train, epochs=1, batch_size=10000)
    y_pred = modelss.predict(x_test)
    yp, yt = [], []
    for i in range(len(y_pred)):
        yp.append(y_pred[i].argmax())
        yt.append(y_test[i].argmax())
    score = accuracy_score(yt, yp)
    print(score)
    nni.report_final_result(score)
    # 主程序
if __name__ == '__main__':
    # 获取数据集
    x_train, y_train, x_test, y_test = load_data()
    # 指定超参数来源
    RECEIVED_PARAMS = nni.get_next_parameter()
    # 获取默认超参数
    PARAMS = get_default_parameters()
    # 参数更新
    PARAMS.update(RECEIVED_PARAMS)
    # 给模型超参数赋值
    model = get_model(PARAMS)
    # 开始超参寻优
    run(x_train, y_train, x_test, y_test, model)

运行结果及报错内容

可以看最后一个图里，succeeded=0,failed=29,还有一个在运行中。右边也没有出现数据

我的解答思路和尝试过的方法

搜了一圈没看过相关错误贴

我想要达到的结果

成功运行