andrewssobral

""" Trains an agent with (stochastic) Policy Gradients on Pong. Uses OpenAI Gym. """
import numpy as np
import cPickle as pickle
import gym

# hyperparameters
H = 200 # number of hidden layer neurons
batch_size = 10 # every how many episodes to do a param update?
learning_rate = 1e-4
gamma = 0.99 # discount factor for reward

andrewssobral

'''This script goes along the blog post
"Building powerful image classification models using very little data"
from blog.keras.io.

It uses data that can be downloaded at:
https://www.kaggle.com/c/dogs-vs-cats/data

In our setup, we:
- created a data/ folder
- created train/ and validation/ subfolders inside data/

andrewssobral

"""
A bare bones examples of optimizing a black-box function (f) using
Natural Evolution Strategies (NES), where the parameter distribution is a 
gaussian of fixed standard deviation.
"""

import numpy as np
np.random.seed(0)

# the function we want to optimize

andrewssobral

Building Tensorflow from source on Linux for maximum performance:

TensorFlow is now distributed under an Apache v2 open source license on GitHub.

Step 1. Install NVIDIA CUDA:

To use TensorFlow with NVIDIA GPUs, the first step is to install the CUDA Toolkit.

Step 2. Install NVIDIA cuDNN:

andrewssobral

import numpy as np
import matplotlib.pyplot as plt
from pylab import *
import os
import sys
import cv2
from PIL import Image
from keras.preprocessing.image import *
from keras.models import load_model
import keras.backend as K

andrewssobral

# batch of files
find . -name "*.php" | while read line; do expand -t 4 $line > $line.new; mv $line.new $line; done

# single file
find CMakeLists.txt | while read line; do expand -t 2 $line > $line.new; mv $line.new $line; done

andrewssobral

ls -1 | gshuf -n 560 | xargs -I '{}' mv '{}' folder/

andrewssobral

# PNG TO JPG
mkdir jpegs
sips -s format jpeg ./*.png --out jpegs

# JPG TO PNG
mkdir pngs
sips -s format png ./*.jpg --out pngs

# GIF TO JPG
find . -iname "*.gif" -type f -exec sh -c 'sips -s format jpeg "$0" --out "${0%.gif}.jpeg"' {} \;

andrewssobral

import numpy as np
import matplotlib.pyplot as plt
from pylab import *
import os
import sys
import pickle
from keras.optimizers import SGD, Adam, Nadam
from keras.callbacks import *
from keras.objectives import *
from keras.metrics import binary_accuracy

andrewssobral

Setup

Client side

Create the container and attach to it:

docker run -itd --name=client_setup ubuntu /bin/bash
docker attach client_setup

Then, inside the container, run: