- 博客(15)
- 资源 (2)
- 收藏
- 关注
RSS订阅原创 机器学习方向python面试题集锦
1 谈谈python 指针变量的理解，并解释python交换两个变量的值实现语句a,b=b,a原理？答：python的变量都是指针，python并不会为定义的变量开辟内存空间，而是为值开辟内存空间。因此，a,b=b,a语句中，a指针指向b的值，b指针指向a的值，从而实现了变量值的交换。2 输出Fibonacci数列，上限100，递归公式为F[n]=F[n-1]+F[n-2]...
2018-12-29 11:57:02
402
原创 训练集、验证集和测试集随机划分实现代码
VALIDATION_PERCENTAGE = 10TEST_PERCENTAGE = 10validation_images=[]validation_labels=[]testing_images=[]testing_labels=[]training_images=[]training_labels=[]chance = np.random.randint(100)i...
2018-12-29 11:38:51
11029
原创 深度学习的归一化和反归一化
医学影像应用场景的特点是训练数据少，数据分布高度不均匀，数据标注的一致性较差，数据类型丰富（多模态，文本+影像等）。1 除最大值法def read_and_normalize_train_data(): train_data, train_label = load_train() print('Convert to numpy...') train_data ...
2018-12-28 17:41:27
11829
1
原创 numpy和matlab思维差异
目录numpy核心对象是多维数组 方括号的深度 numpy数组怎么index 举例：高频函数画图numpy核心对象是多维数组numpy 可以说并没有矩阵的概念，它的核心思维方式是多维数组。>>> np.zeros((256,256)) #产生一个256×256的全0矩阵需要两个括号，行列均需赋值。或>>> np.zeros([256...
2018-12-28 11:15:38
564
原创 keras np_utils()函数
from keras.utils import np_utilsN_CLASSES = 3label = [0,0,0,1,1,1,2,2,2]train_label = np_utils.to_categorical(label, N_CLASSES)train_labelOut[21]:array([[1., 0., 0.], [1., 0., 0.], ...
2018-12-27 17:20:48
9341
原创 python随机打乱代码实现
train_data 和 train_label保持对应state = np.random.get_state()np.random.shuffle(train_data)np.random.set_state(state)np.random.shuffle(train_label)原理numpy.random.set_state(state)Set the internal...
2018-12-27 14:13:58
732
原创 python记时命令
import timestart_time = time.time()......print('Total processing time: {} seconds'.format(round(time.time() - start_time, 2)))输出：Total processing data time: 132.31 seconds或start_time = tim...
2018-12-26 18:11:17
326
原创 High-grade and low-grade tumor
According to American Cancer Society, cancer is typically labelled in stages from I to IV, with IV being the most serious. The grade for most cancers is a measure of how abnormal the cancer cells loo...
2018-12-21 14:25:52
562
原创 tensorflow中list转换为tensor
a = tf.reshape(b, [1, -1])I don't know how many columes it can be, but I only want one row. Ora = tf.reshape(b, [-1, 3])I don't know how many rows it can be, but I want 3 columes.More exam...
2018-12-20 14:04:43
18367
原创 ubuntu18.04+CUDA10+CUDNN+tensorflow_gpu installation guide医疗神经网络工作站搭建
重装系统蔑视
2018-12-17 17:54:47
897
1
原创 python vars()你可能没有彻底理解
先看官方文档vars([object])¶Return the __dict__ attribute for a module, class, instance, or any other object with a __dict__attribute.Objects such as modules and instances have an updateable __dict__ a...
2018-12-05 16:04:56
1362
原创 numpy.ones,numpy.zeros,tf.ones,tf.zeros,np.ones_like,tf.ones_like对比
numpy.ones_like(),numpy.zeros_like()>>> x = np.arange(6)>>> x = x.reshape((2, 3))>>> xarray([[0, 1, 2], [3, 4, 5]])>>> np.ones_like(x)array([[1, 1, 1]...
2018-12-05 12:09:46
1015
Curriculum Learning
2019-01-24
深度学习在生物医学中的机遇与挑战
2018-11-16
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人