I trained the model using CNTK, called simple.dnn now in the testing phase I do not want to install CNTK on windows, but use the trained model with python. How can I use a trained model (weights, ...) for testing with python?
You can use the load_model function, see https://www.cntk.ai/pythondocs/cntk.html?highlight=load_model#cntk.persist.load_model . The main thread should look like this:
load_model
from cntk import load_model loaded_model = load_model("yourModel.model", 'float') output = model.eval(arguments)
Source: https://habr.com/ru/post/1259500/More articles:Android data binding does not work - androidHow to interact between C # and C ++ - c ++Vue.js - Testing asynchronously returned data - unit-testingHow to Determine Incoming Call Dial Phone Number in iOS 10 - iosUID not displayed from constructor - authenticationHow to use setRowViewSelected from ListRowPresenter - androidWhat is the difference between orm and ADO.net? - c #Simple and other times in Chrome developer tools. Why has the browser been inactive for so long? - performanceHow to verify that a specific class method has been passed as a parameter? - javaJava: BaseN universal encoder / decoder working with large data sizes - javaAll Articles