Tim Blog

Himmel oder Hölle

deep learning graph plot

this is intend to introduce some tools to draw deep learning graph beautifully ML visual there is a common ppt from dair.ai community, we can modify and reuse these plots https://docs.google.com/pr...

Posted by neverset on January 4, 2021

reinforcement learning

TF Agents TF agents is a library to run different reinforced learning frameworks based on tensorflow. Official tutorial is here: https://github.com/tensorflow/agents/tree/master/docs/tutorials. Ava...

Posted by neverset on January 4, 2021

face feature detection

DLib a machine learning library for image processing using the C++ providing C/C++, Python, and Java interface. Dlib is able to detect face features and provide a map of landmark points that surrou...

Posted by neverset on January 4, 2021

supported vector machine

Support vector machines (SVM) is a supervised machine learning. Application areas include: classification regression Its biggest advantage is that it can define both a linear or a non-linear d...

Posted by neverset on December 29, 2020

ms word table

python-docx is the library needed for python to interact with ms word installation pip install python-docx usage def describe_text(text): import re, string description = dict() ...

Posted by neverset on December 29, 2020

dimension reduction

Linear Algebra Methods Matrix factorization methods Principal Components Analysis # evaluate pca with logistic regression algorithm for classification from numpy import mean fr...

Posted by neverset on December 29, 2020

chatbot

There are two types of chatbots: Rule-Based chatbot can only response on constrained rules set beforehand Self-Learned chatbot is trained with some neutral networks to react for response NL...

Posted by neverset on December 29, 2020

automl

automl can deploy thousands of models in production, with far less grunt work and hand-tuning, reducing turn-around-time drastically. Auto-Sklearn it implements Bayesian Optimization for searchin...

Posted by neverset on December 29, 2020

web scraping

Tips request with API rather than front-end url #slow solution import requests from bs4 import BeautifulSoup def parsePrice(): r = requests.get("https://finance.yahoo.com/quote/FB?p=FB") soup = B...

Posted by neverset on December 28, 2020

tensorflow serving

define servable TensorFlow function as servable class Adder(tf.Module): @tf.function(input_signature=[tf.TensorSpec(shape=[None,3], dtype=tf.float32, name="x")]) def sum_two(self, x): ...

Posted by neverset on December 27, 2020