Question

I am a window10 user I m running tensorflow in jupyter notebook, perceptron works fine, but cnn kernel crashes. I also tried increasing the buffer size, but the kernel keeps crashing. I made a model with cnn in Kolab, and when I download the model and run it with a Jupyter notebook, the kernel dies. Also, making a model with a jupyter notebook kills the kernel. In jupyter notebook, training with perceptron is possible, but with cnn it always kills the kernel. But .summary() works. What s wrong?

import numpy as np
import tensorflow as tf


testcnn=tf.keras.models.load_model("my_lasthomework_cnn.h5")
testcnn.summary()
import pandas as pd

from PIL import Image
test_dir= dir 
test_csv = pd.read_csv(test_dir+ test.csv )
test_images=[]
for file in test_csv[ data ]:
    image=np.array(Image.open(test_dir+ test/ +file))
    test_images.append(image)
test_images=np.array(test_images)


for i in test_images:
    p1=testcnn.predict(i.reshape(1,64,64,3))
    print(np.argmax(p1))

Answer 1

我正在通过机器/深造课程,并遇到同样的问题——我手提的手提电脑与万国邮联一道在管理示范培训说明书时将坠毁。

借助Windows 11 Preview 的剪辑,在WSL上站出来,花了一些时间寻找问题的根源/解决办法。

然而,上文关于腐败的笔记和将代码移至新笔记(以及单独/新的虚拟环境)的建议,使问题消失。

不过,我必须说的是,我知道失败的根源,而新的笔记本/文告解决了这一问题。

友情链接