QA Official

Keras Method of Saving Model 2019-05-04
use to save the Keras model and weights in an HDF5 file that will contain: model structure to reconstruct the model model weight training configuration (loss function, optimizer, etc.) The state of the optimizer to start where the last training was interrupted. Use Keras. Models. Load _ Model (FilePath) to re-instantiate your model. If the training configuration is stored in the file, the function will also compile the model

Keras functional API 2019-05-04
Example 1: Fully Connected Network from keras.layers import Input, Dense from keras.models import Model # 这部分返回一个张量 inputs = Input(shape=(784,)) # 层的实例是可调用的,它以张量为参数,并且返回一个张量 x = Dense(64, activation='relu')(inputs) x = Dense(64, activation='relu')(x) predictions = Dense(10, activation='softmax')(x) # 这部分

Python Machine Learning Library scikit-learn Practice Multiple Machine Learning Algorithms and Compare Precision ~ (≡≡)/~ 2019-05-04
Python machine learning library scikit-learn practice [email protected]   1. Overview Machine learning algorithms have become "well known" under the influence of big data in recent years. Even if you don't know each algorithm theory and call you one or two famous algorithms, you can blurt it out with your head held high.Of course, although the forest of algorithms is large, there are still limited means.

Python sklearn model selection 2019-05-04
1. The main functions are as follows: 1.classification分类 2.Regression回归 3.Clustering聚类 4.Dimensionality reduction降维 5.Model sele

Python standard machine learning library-introduction to sklearn (1) 2019-05-04
Starting from this part, I mainly share the basic experience and main harvest of learning sklearn library. 1. Basic Usage of Datasets Take the use of handwritten data sets as an example use fromsklearn.datasetimportload _ digits to load datasets parameters return_X_y: the default is false, which means all the information (data and target) of the data is returned in the form of a dictionary;If True, the data is returned as

Python+sklearn completes machine learning tasks 2019-05-04
task Congratulations on your successful internship in a financial company. On your first day at work, you are still excited.At this moment, the supervisor called you over and showed you a document. The contents of the file look like this: The supervisor said this is the company's valuable data asset.I urge you to read carefully and find out the rules from the figures so as to make a wise Foreign currency loans.

Visio Skills Summary 2019-05-04
Visio Skills SummaryWhen drawing the organization chart: file-new-business-organization chart, you can quickly draw what the template needs.1. when Visio draws a picture, when two straight lines cross, there will always be a cross-line mark by default, which is very uncomfortable. the removal method is: select the line, and then menu format-> behavior-> connecting line-> cross-line-> add-> never.2. Add connection points.There is a drop-down button next to the button of the connection line.

keras Chinese Document Note 4-Configuration and Installation 2019-05-04
configuration Recommended Configuration If you are a college student or senior researcher, and your laboratory or personal funds are abundant, it is recommended that you adopt the following configuration: Motherboard: X299 or Z270 Central Processor: i7-6950x or i7-7700K and above memory: brand memory, with a total capacity of more than 32G, which consists of 4 channels or 8 channels according to the motherboard. SSD: brand solid-state disk, with a capacity

keras Chinese Document Notebook 17-Keras as the Thin Interface of tensorflow 2019-05-04
Keras as Thin Interface of tensorflow Call Keras Layer in tensorflow let's start with a simple example: MNIST numerical classification.We will construct a classifier of TensorFlow by stacking Keras' full connection layers. import tensorflow as tf sess = tf.Session() from keras import backend as K K.set_session(sess) Then, we started to build the model with tensorflow: # this placeholder will contain our input digits, as flat vectors img = tf.placeholder(tf.float32, shape=(None, 784)) Keras can accelerate the definition process of the model:

keras How to Save the Model 2019-05-04
use to save the Keras model and weights in an HDF5 file that will contain: 模型的结构,以便重构该模型 模型的权重 训练配置(损失函数,优化器等) 优化器的状态,以便于从上次训练中断的地方开