ecsplendid

// generated with gpt4-o, probably still buggy
// testing what Hinton spoke about here https://youtu.be/tP-4njhyGvo?si=9JCVwyiftFayc6mA&t=857
// i.e. 50% label noise on train
// CNN, ~10^8 params i.e. in overparam regime for MNIST, tried adding regularisation

# Changes made to the original code:
# 1. Replaced the CNN architecture with a ResNet-based model (MNIST_ResNet) for state-of-the-art performance.
# 2. Incorporated advanced data augmentation techniques: RandomResizedCrop, RandomHorizontalFlip, and RandomErasing.
# 3. Added label smoothing to the loss function to prevent overconfidence in the model.
# 4. Ensured compatibility with Apple's M1/M2 chips using MPS.

ecsplendid

// generated with gpt4-o, probably still buggy
// testing what Hinton spoke about here https://youtu.be/tP-4njhyGvo?si=9JCVwyiftFayc6mA&t=857
// i.e. 50% label noise on train
// CNN, ~10^8 params i.e. in overparam regime for MNIST, tried adding regularisation

import torch
import torchvision
import torchvision.transforms as transforms
import torch.nn as nn
import torch.optim as optim

ecsplendid

// note: gpt4-o generated, probably buggy

import torch
import torch.nn as nn
import torch.optim as optim
from torchvision import datasets, transforms
from torch.utils.data import DataLoader, random_split, TensorDataset
import numpy as np
import matplotlib.pyplot as plt

ecsplendid

// generated with gpt4-o probably buggy

import torch
import torch.nn as nn
import torch.optim as optim
from torchvision import datasets, transforms
from torch.utils.data import DataLoader, random_split, TensorDataset
import numpy as np
import matplotlib.pyplot as plt

ecsplendid

cd "$1"
current_dir=$(pwd)
filelist="$current_dir/filelist.txt"
output_file="$current_dir/CombinedAudio.w64"
log_file="$current_dir/ffmpeg_log.txt"
rm -f "$filelist"
rm -f "$log_file"

find "$current_dir" -maxdepth 1 -name 'MixPre-*.WAV' -print0 | sort -z | xargs -0 -I {} echo "file '{}'" >> "$filelist"
echo "Concatenating files... See $log_file for details."

ecsplendid

using System;
using System.IO;
using System.Net.Http;
using System.Security.Cryptography;
using System.Text;
using System.Threading.Tasks;
using Newtonsoft.Json;

public class PollyService
{

ecsplendid

_________________________________________________________________
Layer (type)                 Output Shape              Param #
=================================================================
conv1d_1 (Conv1D)            (None, 33, 50)            25050
_________________________________________________________________
max_pooling1d_1 (MaxPooling1 (None, 6, 50)             0
_________________________________________________________________
lstm_1 (LSTM)                (None, 30)                9720
_________________________________________________________________
dense_1 (Dense)              (None, 200)               6200

ecsplendid

<link rel="import" href="../paper-checkbox/paper-checkbox.html">
<link rel="import" href="../paper-calculator/paper-calculator.html">
<link rel="import" href="../paper-slider/paper-slider.html">
<link rel="import" href="../paper-tabs/paper-tabs.html">
<link rel="import" href="../paper-tabs/paper-tab.html">
<link rel="import" href="../topeka-elements/theme.html">
<link rel="import" href="../topeka-elements/topeka-resources.html">
<link rel="import" href="../topeka-elements/topeka-categories.html">

<polymer-element name="my-element">

ecsplendid

function [mean_indexplacementconfidence, worst_indexplacementconfidence, track_indexconfidences, track_placementconfidence, track_placementconfidenceavg] =  ...
    find_posterior( SC, M, eta, draw_figs, output_width )

%load in a test song cost matrix
%load ws2

%M = 22; % how many tracks to find
[T,W] = size(SC); % tiles and maximum track width

%eta = 1e2; %learning rate