404akhan

import matplotlib.pyplot as plt
import numpy as np


def transform(x):
    return x + np.random.normal(size=len(x)) * 0.1

t = np.linspace(0, 30, 300)  # Time dimension
x1 = np.sin(t)  # X vector values
y1 = np.cos(t)  # Y vector values

404akhan

import numpy as np
import matplotlib.pyplot as plt

np.random.seed(20)


### Set linear regression problem
a_true = 1
b_true = 2

404akhan

import numpy as np
import matplotlib.pyplot as plt


np.random.seed(17)


def loss(x, y):
    ### Default loss. Can modify to others
    return x ** 2 + y ** 2

404akhan

from math import e, sqrt, log, pi
import random
from collections import defaultdict
import math
import matplotlib.pyplot as plt


def f1(q):
    value = (1 + math.erf(q * 0.005)) / 2

404akhan

struct custom_hash {
    static uint64_t splitmix64(uint64_t x) {
        // http://xorshift.di.unimi.it/splitmix64.c
        x += 0x9e3779b97f4a7c15;
        x = (x ^ (x >> 30)) * 0xbf58476d1ce4e5b9;
        x = (x ^ (x >> 27)) * 0x94d049bb133111eb;
        return x ^ (x >> 31);
    }

    size_t operator()(uint64_t x) const {

404akhan

struct Fenw {
	int n;
	vector<ll> bit;

	Fenw(int n): n(n) {
		bit.assign(n + 1, 0);
	}

	void upd(int i, ll v) {
		for(; i <= n; i += i & (-i))

404akhan

PART 4.

We did the testing using JaCoCo. We build the project and then generate JaCoCo report.
Using the following commands.
./gradlew build
./gradlew jacocoTestReport

We have 72 tests testing 9 different classes. All test methods are in diet-bot\src\test\java\com\example\bot\spring folder. 

Our test coverage report folder is attached in the following link:

404akhan

A3C, tensorflow, 42x42 input, 1 day
https://github.com/404akhan/custom-RL/tree/master/a3c

404akhan

import gym
from gym.wrappers import Monitor
import itertools
import numpy as np
import os
import random
import sys
import tensorflow as tf

from collections import deque, namedtuple

404akhan

import torch.nn as nn 
import torch.nn.functional as F


def deconv(c_in, c_out, k_size, stride=2, pad=1, bn=True):
    """Custom deconvolutional layer for simplicity."""
    layers = []
    layers.append(nn.ConvTranspose2d(c_in, c_out, k_size, stride, pad))
    if bn:
        layers.append(nn.BatchNorm2d(c_out, affine=False))