Open qiuyuan666 opened 4 years ago
woodywff
commented
4 years ago 您好，1. n4 bias is not necessary in this project; 不用n4 bias校正就在normalize_image(in_file, out_file, bias_correction=True)中置bias_correction=False; n4 bias不影响数据格式.
qiuyuan666
commented
4 years ago 您好，又来打扰您啦~ 运行train_model.py遇到的问题： 1.运行到第152行，get_training_and_validation_generators( )函数时会进入generator.py执行该函数，执行完毕后发现读取的datafile是生成的data"文件夹下的data.h5文件，显示training_list, validation_list，num_patches_training，num_patches_val都是0，这是因为根本没有读取到数据吧？ 在运行之前没有单独运行process.py，没有做预处理将数据存在对应文件夹，所以generator.py里的函数就没读到数据吗？
2.继续运行到第177行进入train_model( )函数后就一直在这个函数里不动了，具体显示如下：
WARNING:tensorflow:From /home/anaconda3/envs/tensorflow/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.
Instructions for updating:
Please use rate instead of keep_prob. Rate should be set to rate = 1 - keep_prob.
Backend TkAgg is interactive backend. Turning interactive mode on.
Loading previous validation split...
Number of training steps in each epoch: 1
Number of validation steps in each epoch: 1
WARNING:tensorflow:From /home/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.
Epoch 1/300
这里一直等了一天一夜都没有反应，请问这是为啥？
3.运行train_model.py后生成了几个文件分别是：data.h5/list_cv1_train.pkl/list_cv1_val.pkl/mean_std.pkl/survival_data.csv，这几个文件分别是什么？
期待您的解答【抱拳.jpg】
qiuyuan666
commented
4 years ago 您好，我运行了预处理的代码，预处理后的数据存放目录和您的original_tree对比后是一样的。然后我重新运行了train_model.py，但是还是training_list, validation_list，num_patches_training，num_patches_val都是0，运行后结果如下：
‘’‘
totalMemory: 7.79GiB freeMemory: 7.35GiB
2020-10-09 15:34:57.603044: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1512] Adding visible gpu devices: 0
2020-10-09 15:34:57.604181: I tensorflow/core/common_runtime/gpu/gpu_device.cc:984] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-10-09 15:34:57.604200: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 0
2020-10-09 15:34:57.604211: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1003] 0: N
2020-10-09 15:34:57.604299: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device
WARNING:tensorflow:From /home/anaconda3/envs/tensorflow/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.
Instructions for updating:
Please use rate instead of keep_prob. Rate should be set to rate = 1 - keep_prob.
Loading previous validation split...
Number of training steps in each epoch: 1
Number of validation steps in each epoch: 1
WARNING:tensorflow:From /home/qiuyuan/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.
Epoch 1/300
Process finished with exit code 139(interrupted by signal 11:SIGSEGV） ’‘’ (1)想问一下这里为什么一直读取不到数据呢？可以指点一下吗？
感谢您开源您们的代码，我是先读了您的文章才找到这个开源代码的，我想用您的代码来实现MRI分割（数据集是brats2019的训练数据和验证数据）。 想问您几个问题： 1.先运行preprocess.py对数据集进行n4 bias校正和归一化处理是么？我如果不进行n4 bias 校正，要重新设置哪些参数？而且不管是否进行n4 bias 校正，预处理后的数据格式都是一样的吗？ 2.接下来运行run_validation.py，再运行evaluate.py进行评价吗？ 3.能否注释一下每个.py文件的功能？
问您这么多问题，还望您见谅，也麻烦您可以解答一下