Chris-hughes10

Chris-hughes10

# this example is taken from 
# https://github.com/Chris-hughes10/pytorch-accelerated/blob/main/examples/train_mnist.py

import os

from torch import nn, optim
from torch.utils.data import random_split
from torchvision import transforms
from torchvision.datasets import MNIST

Chris-hughes10

from torchmetrics import MetricCollection, Accuracy, Precision, Recall

class TrainerWithMetrics(Trainer):
    def __init__(self, num_classes, *args, **kwargs):
        super().__init__(*args, **kwargs)

        # this will be moved to the correct device automatically by the
        # MoveModulesToDeviceCallback callback, which is used by default
        self.metrics = MetricCollection(
            {

Chris-hughes10

# https://github.com/Chris-hughes10/pytorch-accelerated/blob/main/examples/metrics/train_with_metrics_in_loop.py
import os

from torch import nn, optim
from torch.utils.data import random_split
from torchmetrics import MetricCollection, Accuracy, Precision, Recall
from torchvision import transforms
from torchvision.datasets import MNIST

from pytorch_accelerated import Trainer

Chris-hughes10

from torchmetrics import MetricCollection, Accuracy, Precision, Recall

from pytorch_accelerated.callbacks import TrainerCallback


class ClassificationMetricsCallback(TrainerCallback):
    def __init__(self, num_classes):
        self.metrics = MetricCollection(
            {
                "accuracy": Accuracy(num_classes=num_classes),

Chris-hughes10

# https://github.com/Chris-hughes10/pytorch-accelerated/blob/main/examples/metrics/train_with_metrics_in_callback.py

import os

from torch import nn, optim
from torch.utils.data import random_split
from torchmetrics import MetricCollection, Accuracy, Precision, Recall
from torchvision import transforms
from torchvision.datasets import MNIST

Chris-hughes10

import torch
from torch import nn

class MfDotBias(nn.Module):

    def __init__(
        self, n_factors, n_users, n_items, ratings_range=None, use_biases=True
    ):
        super().__init__()
        self.bias = use_biases

Chris-hughes10

from pytorch_accelerated.callbacks import TrainerCallback
import torchmetrics

class RecommenderMetricsCallback(TrainerCallback):
    def __init__(self):
        self.metrics = torchmetrics.MetricCollection(
            {
                "mse": torchmetrics.MeanSquaredError(),
                "mae": torchmetrics.MeanAbsoluteError(),
            }

Chris-hughes10

Chris-hughes10

from pathlib import Path

from timm.data.parsers.parser import Parser


class ParserImageName(Parser):
    def __init__(self, root, class_to_idx=None):
        super().__init__()

        self.root = Path(root)