Anjum48

sudo apt update && sudo apt upgrade -y

sudo apt -y install build-essential git zsh python3-pip gnome-shell-extension-manager openssh-server curl nodejs htop mdadm ffmpeg lm-sensors npm tree

sudo apt -y install linux-headers-$(uname -r)
sudo apt-key del 7fa2af80

# Latest driver download
# https://www.nvidia.co.uk/download/driverResults.aspx/205541/en-uk
wget https://uk.download.nvidia.com/XFree86/Linux-x86_64/535.54.03/NVIDIA-Linux-x86_64-535.54.03.run

Anjum48

import numpy as np
import matplotlib.pyplot as plt
import random
import matplotlib.animation as animation
from matplotlib import cm

# s curve function
def _f(a,b,c,t):
    return c*1/(1+a*np.exp(-b*t))
# function for s curve with added noise

Anjum48

class HardMiningBatchSampler(BatchSampler):
    """
    Creates batches that only contain a class once and and chosen based on embedding distance.
    Used for NPairLoss
    """

    def __init__(self, labels, batch_size, drop_last=True, classes=None):
        self.labels = labels
        self.batch_size = batch_size
        # self.C = batch_size * 2 if classes is None else classes

Anjum48

from category_encoders import utils
from sklearn.base import BaseEstimator, TransformerMixin
from sklearn.model_selection import StratifiedKFold
import category_encoders as encoders
import pandas as pd
import numpy as np
import copy


class NestedCVWrapper(BaseEstimator, TransformerMixin):

Anjum48

import numpy as np
import pandas as pd
import os
import matplotlib.pyplot as plt

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.utils.data import Dataset, DataLoader

Anjum48

"""
Implementation of DDPG - Deep Deterministic Policy Gradient
Algorithm and hyperparameter details can be found here: http://arxiv.org/pdf/1509.02971v2.pdf
Variance scaling paper: https://arxiv.org/pdf/1502.01852v1.pdf
Thanks to GitHub users yanpanlau, pemami4911, songrotek and JunhongXu for their DDPG examples

Batch normalisation on the actor accelerates learning but has poor long term stability. Applying to the critic breaks
it, particularly on the state branch. Not sure why but I think this issue is specific to this environment 
"""
import numpy as np