halflearned

from linearmodels import PanelOLS
import numpy as np
import pandas as pd
from itertools import product
import seaborn as sns
import matplotlib.pyplot as plt

def generate_data(N, T, rho, sigma_a = 5, sigma_eps=1, burn_in = 100):
    # dumb for-loop to generate data, not efficient but easy to understand
    data = []

halflearned

class VGGSoundDADAPolicy(nn.Module):
    """ Transformation policy learned after applying DADA algorithm to VGG-Sound """
    
    def __init__(self, num_time_subpolicies=2, num_freq_subpolicies=2):
        super().__init__()
        self.num_time_subpolicies = num_time_subpolicies
        self.num_freq_subpolicies = num_freq_subpolicies
        
    def forward(self, x):
        time_subpolicy_indices = np.random.choice(

halflearned

import os
import boto3
import numpy as np
import torch
import torchvision
import torchvision.transforms as transforms
from torch.utils.data import Subset

def upload_to_s3(local_path, s3_path):
    s3 = boto3.client('s3', region_name="us-west-2")

halflearned

import argparse
import csv
import json
import subprocess
from multiprocessing import Pool
from os import remove, rename, makedirs
from os.path import exists, getsize
from time import sleep

halflearned

import numpy as np
from scipy.stats import norm, expon

def quantile_with_ci(x, q, alpha=.05, num_boot=10_000):
    n = len(x)
    quantile_estimate = np.quantile(x, q)
    idx_boot = np.random.randint(n, size=(num_boot, n))
    quantile_boot = np.quantile(x[idx_boot], q=q, axis=1)
    quantile_se = np.std(quantile_boot)
    half_ci = norm.ppf(1 - alpha/2) * quantile_se

halflearned

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

df = pd.read_excel("/Users/halflearned/Downloads/qPCR.xlsx")

# Rename expression column to something sensible
df = df.rename(columns={"2^(-ΔΔCq)": "expression"})

# Create the column that will be our horizontal axis

halflearned

import pickle
from datetime import datetime, timezone
from itertools import chain

import numpy as np
import pytest

from sleep_x_ml_reconciliation_service import SleepXMLReconciliation
from sleep_x_ml_reconciliation_service.constants import (
    AWAKE_STAGE,

halflearned

library(purrr)
library(Matrix)
library(sandwich)
library(MASS)



romano_wolf_correction <- function(t.orig, t.boot) {
  # See http://ftp.iza.org/dp12845.pdf page 8
  abs.t.orig <- abs(t.orig)

halflearned

import pytest
import pickle
from itertools import chain
from datetime import datetime, timezone

from sleep_x_ml_reconciliation_service import SleepXMLReconciliation
from sleep_x_ml_reconciliation_service.constants.reconciliation import (
    UNINTENDED_USER_FLAG,
    LARGE_SLEEP_GAP_FLAG,
    PUT_GAP_TOLERANCE_SECONDS

halflearned

# WIP: This code does NOT work

import numpy as np
import matplotlib.pyplot as plt
from scipy.special import beta
from scipy.optimize import minimize


def neg_beta_log_likelihood(z, params):
    a0, b0, a1, b1, tau = params