rbrigden

from __future__ import division, print_function, absolute_import

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

# Import MNIST data
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data", one_hot=False)

rbrigden

import matplotlib
import matplotlib.pyplot as plt
import numpy as np
import random

class TenArmedBandit(object):
    
    def __init__(self):
        self.action_space = 10
        self.q_true = np.random.randn(self.action_space)  

rbrigden

#!/bin/bash

cd ~

apt-get install git

# setup dotfiles
rm -rf .vim*
git clone https://github.com/rbrigden/dotfiles.git .dotfiles
ln -s .dotfiles/.vimrc .vimrc

rbrigden

import numpy as np
import os

class WSJ():
    """ Load the WSJ speech dataset
        
        Ensure WSJ_PATH is path to directory containing 
        all data files (.npy) provided on Kaggle.
        
        Example usage:

rbrigden

import numpy as np
import copy

# NOTE: a = 1 is a(-), a = 0 is a(+)
gamma = 0.9
action_space = 2
state_space = 4
eps = 1e-9

# Action conditioned reward function

rbrigden

import torch
import torch.nn as nn
from torch.autograd import Variable

affine = nn.Linear(10, 10)
# A linear mapping to a random vector... just for quick demo purposes
x = Variable(torch.randn(100, 10))
y = Variable(torch.randn(100, 10))

weird_loss = torch.mean(torch.exp(affine.weight))

rbrigden

import torch
import torch.nn as nn
import torch.nn.functional as F
import numpy as np

class MLP(nn.Module):

    def __init__(self, input_size, feature_categories):
        super(MLP, self).__init__()
        self.feature_categories = feature_categories