Skip to content

Instantly share code, notes, and snippets.

Mirodil Mirodil

View GitHub Profile
Mirodil /
Created Dec 10, 2018
PyTorch Learning Rate Finder
def find_learning_rate(model, data_loader, criterion, lr:tuple=(1e-7, 1), epochs:int=1):
history = []
min_lr, max_lr = lr
num_batches = epochs * len(data_loader)
last_avg_loss, i, beta = 0, 0, 0.98
# preserve initial state
initial_weights = './temp.model', initial_weights)
Mirodil /
Last active Dec 14, 2018
LearningRateFinder for keras
class LearningRateFinder(Callback):
This callback implements a learning rate finder(LRF)
The learning rate is constantly increased during training.
On training end, the training loss is plotted against the learning rate.
One may choose a learning rate for a model based on the given graph,
selecting a value slightly before the minimal training loss.
# Example
lrf = LearningRateFinder([0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05]), y_train, epochs=1, batch_size=128, callbacks=[lrf])
Mirodil /
Last active May 10, 2018
FileListIterator for keras
import numpy as np
from keras import backend as K
from keras.preprocessing.image import Iterator, load_img, img_to_array
class FileListIterator(Iterator):
"""Iterator capable of reading images from an array of the filenames.
# Arguments
filenames: Path to the directory to read images from.
Each subdirectory in this directory will be
considered to contain images from one class,
Mirodil / config.js
Last active Jun 19, 2017
AngularJs: OAuth 2.0 Refresh Token
View config.js
angular.module('app', [])
.config(['$httpProvider', function($httpProvider) {
$httpProvider.interceptors.push(['$q', '$window', '$timeout', '$injector', function ($q, $window, $timeout, $injector) {
var $login, $http, $auth;
$timeout(function () {
$login = $injector.get('$login');
$http = $injector.get('$http');
$auth = $injector.get('$auth');
You can’t perform that action at this time.