Title: Ray Tune을 사용한 하이퍼파라미터 튜닝 — 파이토치 한국어 튜토리얼 (PyTorch tutorials in Korean)
Open Graph Title: Ray Tune을 사용한 하이퍼파라미터 튜닝
Description: 번역: 심형준 하이퍼파라미터 튜닝은 보통의 모델과 매우 정확한 모델간의 차이를 만들어 낼 수 있습니다. 종종 다른 학습률(Learnig rate)을 선택하거나 layer size를 변경하는 것과 같은 간단한 작업만으로도 모델 성능에 큰 영향을 미치기도 합니다. 다행히, 최적의 매개변수 조합을 찾는데 도움이 되는 도구가 있습니다. Ray Tune 은 분산 하이퍼파라미터 튜닝을 위한 업계 표준 도구입니다. Ray Tune은 최신 하이퍼파라미터 검색 알고리즘을 포함하고 다양한 분석 라이브러리와 통합되며 기본적으로 Ray 의 분산 기...
Open Graph Description: 번역: 심형준 하이퍼파라미터 튜닝은 보통의 모델과 매우 정확한 모델간의 차이를 만들어 낼 수 있습니다. 종종 다른 학습률(Learnig rate)을 선택하거나 layer size를 변경하는 것과 같은 간단한 작업만으로도 모델 성능에 큰 영향을 미치기도 합니다. 다행히, 최적의 매개변수 조합을 찾는데 도움이 되는 도구가 있습니다. Ray Tune 은 분산 하이퍼파라미터 튜닝을 위한 업계 표준 도구입니다. Ray Tune은 최신 하이퍼파라미터 검색 알고리즘을 포함하고 다양한 분석 라이브러리와 통합되며 기본적으로 Ray 의 분산 기...
Opengraph URL: https://tutorials.pytorch.kr/beginner/hyperparameter_tuning_tutorial.html
Domain: tutorials.pytorch.kr
{
"@context": "https://schema.org",
"@type": "Article",
"name": "Ray Tune\uc744 \uc0ac\uc6a9\ud55c \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd",
"headline": "Ray Tune\uc744 \uc0ac\uc6a9\ud55c \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd",
"description": "PyTorch Documentation. Explore PyTorch, an open-source machine learning library that accelerates the path from research prototyping to production deployment. Discover tutorials, API references, and guides to help you build and deploy deep learning models efficiently.",
"url": "/beginner/hyperparameter_tuning_tutorial.html",
"articleBody": "\ucc38\uace0 Go to the end to download the full example code. Ray Tune\uc744 \uc0ac\uc6a9\ud55c \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd# \ubc88\uc5ed: \uc2ec\ud615\uc900 \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd\uc740 \ubcf4\ud1b5\uc758 \ubaa8\ub378\uacfc \ub9e4\uc6b0 \uc815\ud655\ud55c \ubaa8\ub378\uac04\uc758 \ucc28\uc774\ub97c \ub9cc\ub4e4\uc5b4 \ub0bc \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc885\uc885 \ub2e4\ub978 \ud559\uc2b5\ub960(Learnig rate)\uc744 \uc120\ud0dd\ud558\uac70\ub098 layer size\ub97c \ubcc0\uacbd\ud558\ub294 \uac83\uacfc \uac19\uc740 \uac04\ub2e8\ud55c \uc791\uc5c5\ub9cc\uc73c\ub85c\ub3c4 \ubaa8\ub378 \uc131\ub2a5\uc5d0 \ud070 \uc601\ud5a5\uc744 \ubbf8\uce58\uae30\ub3c4 \ud569\ub2c8\ub2e4. \ub2e4\ud589\ud788, \ucd5c\uc801\uc758 \ub9e4\uac1c\ubcc0\uc218 \uc870\ud569\uc744 \ucc3e\ub294\ub370 \ub3c4\uc6c0\uc774 \ub418\ub294 \ub3c4\uad6c\uac00 \uc788\uc2b5\ub2c8\ub2e4. Ray Tune \uc740 \ubd84\uc0b0 \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd\uc744 \uc704\ud55c \uc5c5\uacc4 \ud45c\uc900 \ub3c4\uad6c\uc785\ub2c8\ub2e4. Ray Tune\uc740 \ucd5c\uc2e0 \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \uac80\uc0c9 \uc54c\uace0\ub9ac\uc998\uc744 \ud3ec\ud568\ud558\uace0 \ub2e4\uc591\ud55c \ubd84\uc11d \ub77c\uc774\ube0c\ub7ec\ub9ac\uc640 \ud1b5\ud569\ub418\uba70 \uae30\ubcf8\uc801\uc73c\ub85c Ray \uc758 \ubd84\uc0b0 \uae30\uacc4 \ud559\uc2b5 \uc5d4\uc9c4 \uc744 \ud1b5\ud574 \ud559\uc2b5\uc744 \uc9c0\uc6d0\ud569\ub2c8\ub2e4. \uc774 \ud29c\ud1a0\ub9ac\uc5bc\uc740 Ray Tune\uc744 \ud30c\uc774\ud1a0\uce58 \ud559\uc2b5 workflow\uc5d0 \ud1b5\ud569\ud558\ub294 \ubc29\ubc95\uc744 \uc54c\ub824\uc90d\ub2c8\ub2e4. CIFAR10 \uc774\ubbf8\uc9c0 \ubd84\ub958\uae30\ub97c \ud6c8\ub828\ud558\uae30 \uc704\ud574 \ud30c\uc774\ud1a0\uce58 \ubb38\uc11c\uc5d0\uc11c \uc774 \ud29c\ud1a0\ub9ac\uc5bc\uc744 \ud655\uc7a5\ud560 \uac83\uc785\ub2c8\ub2e4. \uc544\ub798\uc640 \uac19\uc774 \uc57d\uac04\uc758 \uc218\uc815\ub9cc \ucd94\uac00\ud558\uba74 \ub429\ub2c8\ub2e4. \ud568\uc218\uc5d0\uc11c \ub370\uc774\ud130 \ub85c\ub529 \ubc0f \ud559\uc2b5 \ubd80\ubd84\uc744 \uac10\uc2f8\ub450\uace0, \uc77c\ubd80 \ub124\ud2b8\uc6cc\ud06c \ud30c\ub77c\ubbf8\ud130\ub97c \uad6c\uc131 \uac00\ub2a5\ud558\uac8c \ud558\uace0, \uccb4\ud06c\ud3ec\uc778\ud2b8\ub97c \ucd94\uac00\ud558\uace0 (\uc120\ud0dd \uc0ac\ud56d), \ubaa8\ub378 \ud29c\ub2dd\uc744 \uc704\ud55c \uac80\uc0c9 \uacf5\uac04\uc744 \uc815\uc758\ud569\ub2c8\ub2e4. \uc774 \ud29c\ud1a0\ub9ac\uc5bc\uc744 \uc2e4\ud589\ud558\uae30 \uc704\ud574 \uc544\ub798\uc758 \ud328\ud0a4\uc9c0\uac00 \uc124\uce58\ub418\uc5b4 \uc788\ub294\uc9c0 \ud655\uc778\ud558\uc138\uc694: ray[tune]: \ubc30\ud3ec\ub41c \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd \ub77c\uc774\ube0c\ub7ec\ub9ac torchvision: \ub370\uc774\ud130 \ubcc0\ud615\uc744 \uc704\ud574 \ud544\uc694 \uc124\uc815 / \ubd88\ub7ec\uc624\uae30# \ud544\uc694\ud55c \ub77c\uc774\ube0c\ub7ec\ub9ac\ub4e4\uc744 \ubd88\ub7ec\uc624\ub294 \uac83(import)\uc73c\ub85c \uc2dc\uc791\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4: from functools import partial import os import tempfile from pathlib import Path import torch import torch.nn as nn import torch.nn.functional as F import torch.optim as optim from torch.utils.data import random_split import torchvision import torchvision.transforms as transforms from ray import tune from ray import train from ray.train import Checkpoint, get_checkpoint from ray.tune.schedulers import ASHAScheduler import ray.cloudpickle as pickle /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. \ub300\ubd80\ubd84\uc758 import\ub4e4\uc740 \ud30c\uc774\ud1a0\uce58 \ubaa8\ub378\uc744 \ube4c\ub4dc\ud558\ub294\ub370 \ud544\uc694\ud569\ub2c8\ub2e4. \uac00\uc7a5 \ub9c8\uc9c0\ub9c9\uc758 import\ub9cc\uc774 Ray Tune\uc744 \uc0ac\uc6a9\ud558\uae30 \uc704\ud55c \uac83\uc785\ub2c8\ub2e4. Data loaders# data loader\ub97c \uc790\uccb4 \ud568\uc218\ub85c \uac10\uc2f8\ub450\uace0 \uc804\uc5ed \ub370\uc774\ud130 \ub514\ub809\ud1a0\ub9ac\ub85c \uc804\ub2ec\ud569\ub2c8\ub2e4. \uc774\ub7f0 \uc2dd\uc73c\ub85c \uc11c\ub85c \ub2e4\ub978 \uc2e4\ud5d8\ub4e4 \uac04\uc5d0 \ub370\uc774\ud130 \ub514\ub809\ud1a0\ub9ac\ub97c \uacf5\uc720\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. def load_data(data_dir=\"./data\"): transform = transforms.Compose( [transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))] ) trainset = torchvision.datasets.CIFAR10( root=data_dir, train=True, download=True, transform=transform ) testset = torchvision.datasets.CIFAR10( root=data_dir, train=False, download=True, transform=transform ) return trainset, testset \uad6c\uc131 \uac00\ub2a5\ud55c \uc2e0\uacbd\ub9dd# \uad6c\uc131 \uac00\ub2a5\ud55c \ud30c\ub77c\ubbf8\ud130\ub9cc \ud29c\ub2dd\uc774 \uac00\ub2a5\ud569\ub2c8\ub2e4. \uc774 \uc608\uc2dc\ub97c \ud1b5\ud574 fully connected layer \ud06c\uae30\ub97c \uc9c0\uc815\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4: class Net(nn.Module): def __init__(self, l1=120, l2=84): super(Net, self).__init__() self.conv1 = nn.Conv2d(3, 6, 5) self.pool = nn.MaxPool2d(2, 2) self.conv2 = nn.Conv2d(6, 16, 5) self.fc1 = nn.Linear(16 * 5 * 5, l1) self.fc2 = nn.Linear(l1, l2) self.fc3 = nn.Linear(l2, 10) def forward(self, x): x = self.pool(F.relu(self.conv1(x))) x = self.pool(F.relu(self.conv2(x))) x = torch.flatten(x, 1) # \ubc30\uce58(batch) \ucc28\uc6d0\uc744 \uc81c\uc678\ud55c \ubaa8\ub4e0 \ucc28\uc6d0\uc744 \ud3c9\ud0c4\ud654(flatten) x = F.relu(self.fc1(x)) x = F.relu(self.fc2(x)) x = self.fc3(x) return x \ud559\uc2b5 \ud568\uc218# \ud765\ubbf8\ub97c \ub354\ud574\ubcf4\uace0\uc790 \ud30c\uc774\ud1a0\uce58 \ubb38\uc11c\uc758 \uc608\uc81c \uc77c\ubd80\ub97c \ubcc0\uacbd\ud558\uc5ec \uc18c\uac1c\ud569\ub2c8\ub2e4. \ud559\uc2b5 \uc2a4\ud06c\ub9bd\ud2b8\ub97c train_cifar(config, data_dir=None) \ud568\uc218\ub85c \uac10\uc2f8\ub461\ub2c8\ub2e4. config \ub9e4\uac1c\ubcc0\uc218\ub294 \ud559\uc2b5\ud560 \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130(hyperparameter)\ub97c \ubc1b\uc2b5\ub2c8\ub2e4. data_dir \uc740 \uc5ec\ub7ec \ubc88\uc758 \uc2e4\ud589(run) \uc2dc \ub3d9\uc77c\ud55c \ub370\uc774\ud130 \uc18c\uc2a4\ub97c \uacf5\uc720\ud560 \uc218 \uc788\ub3c4\ub85d \ub370\uc774\ud130\ub97c \uc77d\uace0 \uc800\uc7a5\ud558\ub294 \ub514\ub809\ud1a0\ub9ac\ub97c \uc9c0\uc815\ud569\ub2c8\ub2e4. \ub610\ud55c, checkpoint\uac00 \uc9c0\uc815\ub418\ub294 \uacbd\uc6b0\uc5d0\ub294 \uc2e4\ud589 \uc2dc\uc791 \uc2dc\uc810\uc758 \ubaa8\ub378\uacfc \uc635\ud2f0\ub9c8\uc774\uc800 \uc0c1\ud0dc(optimizer state)\ub97c \ubd88\ub7ec\uc62c \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774 \ud29c\ud1a0\ub9ac\uc5bc\uc758 \uc544\ub798\ucabd\uc5d0\uc11c \uccb4\ud06c\ud3ec\uc778\ud2b8(checkpoint)\ub97c \uc9c0\uc815\ud558\ub294 \ubc29\ubc95\uacfc \uccb4\ud06c\ud3ec\uc778\ud2b8\uc758 \uc6a9\ub3c4\uc5d0 \ub300\ud55c \uc815\ubcf4\ub97c \ud655\uc778\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. net = Net(config[\"l1\"], config[\"l2\"]) checkpoint = get_checkpoint() if checkpoint: with checkpoint.as_directory() as checkpoint_dir: data_path = Path(checkpoint_dir) / \"data.pkl\" with open(data_path, \"rb\") as fp: checkpoint_state = pickle.load(fp) start_epoch = checkpoint_state[\"epoch\"] net.load_state_dict(checkpoint_state[\"net_state_dict\"]) optimizer.load_state_dict(checkpoint_state[\"optimizer_state_dict\"]) else: start_epoch = 0 \ub610\ud55c, \uc635\ud2f0\ub9c8\uc774\uc800\uc758 \ud559\uc2b5\ub960(learning rate)\uc744 \uad6c\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. optimizer = optim.SGD(net.parameters(), lr=config[\"lr\"], momentum=0.9) \ub610\ud55c \ud559\uc2b5 \ub370\uc774\ud130\ub97c \ud559\uc2b5 \ubc0f \uac80\uc99d \uc138\ud2b8\ub85c \ub098\ub215\ub2c8\ub2e4. \ub530\ub77c\uc11c \ub370\uc774\ud130\uc758 80%\ub294 \ubaa8\ub378 \ud559\uc2b5\uc5d0 \uc0ac\uc6a9\ud558\uace0, \ub098\uba38\uc9c0 20%\uc5d0 \ub300\ud574 \uc720\ud6a8\uc131 \uac80\uc0ac \ubc0f \uc190\uc2e4\uc744 \uacc4\uc0b0\ud569\ub2c8\ub2e4. \ud559\uc2b5 \ubc0f \ud14c\uc2a4\ud2b8 \uc138\ud2b8\ub97c \ubc18\ubcf5\ud558\ub294 \ubc30\uce58 \ud06c\uae30\ub3c4 \uad6c\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. DataParallel\uc744 \uc774\uc6a9\ud55c GPU(\ub2e4\uc911)\uc9c0\uc6d0 \ucd94\uac00# \uc774\ubbf8\uc9c0 \ubd84\ub958\ub294 GPU\ub97c \uc0ac\uc6a9\ud560 \ub54c \uc774\uc810\uc774 \ub9ce\uc2b5\ub2c8\ub2e4. \uc6b4\uc88b\uac8c\ub3c4 Ray Tune\uc5d0\uc11c \ud30c\uc774\ud1a0\uce58\uc758 \ucd94\uc0c1\ud654\ub97c \uacc4\uc18d \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \ub530\ub77c\uc11c \uc5ec\ub7ec GPU\uc5d0\uc11c \ub370\uc774\ud130 \ubcd1\ub82c \ud6c8\ub828\uc744 \uc9c0\uc6d0\ud558\uae30 \uc704\ud574 \ubaa8\ub378\uc744 nn.DataParallel \uc73c\ub85c \uac10\uc300 \uc218 \uc788\uc2b5\ub2c8\ub2e4. device = \"cpu\" if torch.cuda.is_available(): device = \"cuda:0\" if torch.cuda.device_count() \u003e 1: net = nn.DataParallel(net) net.to(device) device \ubcc0\uc218\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc0ac\uc6a9 \uac00\ub2a5\ud55c GPU\uac00 \uc5c6\uc744 \ub54c\ub3c4 \ud559\uc2b5\uc774 \uac00\ub2a5\ud55c\uc9c0 \ud655\uc778\ud569\ub2c8\ub2e4. \ud30c\uc774\ud1a0\uce58\ub294 \ub2e4\uc74c\uacfc \uac19\uc774 \ub370\uc774\ud130\ub97c GPU\uba54\ubaa8\ub9ac\uc5d0 \uba85\uc2dc\uc801\uc73c\ub85c \ubcf4\ub0b4\ub3c4\ub85d \uc694\uad6c\ud569\ub2c8\ub2e4. for i, data in enumerate(trainloader, 0): inputs, labels = data inputs, labels = inputs.to(device), labels.to(device) \uc774 \ucf54\ub4dc\ub294 \uc774\uc81c CPU\ub4e4, \ub2e8\uc77c GPU \ubc0f \ub2e4\uc911 GPU\uc5d0 \ub300\ud55c \ud559\uc2b5\uc744 \uc9c0\uc6d0\ud569\ub2c8\ub2e4. \ud2b9\ud788 Ray\ub294 fractional-GPUs \ub3c4 \uc9c0\uc6d0\ud558\ubbc0\ub85c \ubaa8\ub378\uc774 GPU \uba54\ubaa8\ub9ac\uc5d0 \uc801\ud569\ud55c \uc0c1\ud669\uc5d0\uc11c\ub294 \ud14c\uc2a4\ud2b8 \uac04\uc5d0 GPU\ub97c \uacf5\uc720\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub294 \ub098\uc911\uc5d0 \ub2e4\ub8f0 \uac83\uc785\ub2c8\ub2e4. Ray Tune\uc73c\ub85c \ud1b5\uc2e0\ud558\uae30# \uac00\uc7a5 \ud765\ubbf8\ub85c\uc6b4 \ubd80\ubd84\uc740 Ray Tune\uacfc\uc758 \ud1b5\uc2e0\uc785\ub2c8\ub2e4: checkpoint_data = { \"epoch\": epoch, \"net_state_dict\": net.state_dict(), \"optimizer_state_dict\": optimizer.state_dict(), } with tempfile.TemporaryDirectory() as checkpoint_dir: data_path = Path(checkpoint_dir) / \"data.pkl\" with open(data_path, \"wb\") as fp: pickle.dump(checkpoint_data, fp) checkpoint = Checkpoint.from_directory(checkpoint_dir) train.report( {\"loss\": val_loss / val_steps, \"accuracy\": correct / total}, checkpoint=checkpoint, ) \uc5ec\uae30\uc11c \uba3c\uc800 \uccb4\ud06c\ud3ec\uc778\ud2b8\ub97c \uc800\uc7a5\ud55c \ub2e4\uc74c \uc77c\ubd80 \uba54\ud2b8\ub9ad\uc744 Ray Tune\uc5d0 \ub2e4\uc2dc \ubcf4\ub0c5\ub2c8\ub2e4. \ud2b9\ud788, validation loss\uc640 accuracy\ub97c Ray Tune\uc73c\ub85c \ub2e4\uc2dc \ubcf4\ub0c5\ub2c8\ub2e4. \uadf8 \ud6c4 Ray Tune\uc740 \uc774\ub7ec\ud55c \uba54\ud2b8\ub9ad\uc744 \uc0ac\uc6a9\ud558\uc5ec \ucd5c\uc0c1\uc758 \uacb0\uacfc\ub97c \uc720\ub3c4\ud558\ub294 \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \uad6c\uc131\uc744 \uacb0\uc815\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub7ec\ud55c \uba54\ud2b8\ub9ad\ub4e4\uc740 \ub610\ud55c \ub9ac\uc18c\uc2a4 \ub0ad\ube44\ub97c \ubc29\uc9c0\ud558\uae30 \uc704\ud574 \uc131\ub2a5\uc774 \uc88b\uc9c0 \uc54a\uc740 \uc2e4\ud5d8\uc744 \uc870\uae30\uc5d0 \uc911\uc9c0\ud558\ub294 \ub370 \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uccb4\ud06c\ud3ec\uc778\ud2b8 \uc800\uc7a5\uc740 \uc120\ud0dd\uc0ac\ud56d\uc774\uc9c0\ub9cc, Population Based Training \uacfc \uac19\uc740 \uace0\uae09 \uc2a4\ucf00\uc904\ub7ec\ub97c \uc0ac\uc6a9\ud558\uae30 \uc704\ud574\uc11c\ub294 \ud544\uc694\ud569\ub2c8\ub2e4. \ub610\ud55c, \uccb4\ud06c\ud3ec\uc778\ud2b8\ub97c \uc800\uc7a5\ud574\ub450\uba74 \ub098\uc911\uc5d0 \ud559\uc2b5\ub41c \ubaa8\ub378\uc744 \ub85c\ub4dc\ud558\uace0 \ud3c9\uac00 \uc138\ud2b8(test set)\uc5d0\uc11c \uac80\uc99d\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc804\uccb4 \ud559\uc2b5 \ud568\uc218# \uc804\uccb4 \uc608\uc81c \ucf54\ub4dc\ub294 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4. def train_cifar(config, data_dir=None): net = Net(config[\"l1\"], config[\"l2\"]) device = \"cpu\" if torch.cuda.is_available(): device = \"cuda:0\" if torch.cuda.device_count() \u003e 1: net = nn.DataParallel(net) net.to(device) criterion = nn.CrossEntropyLoss() optimizer = optim.SGD(net.parameters(), lr=config[\"lr\"], momentum=0.9) checkpoint = get_checkpoint() if checkpoint: with checkpoint.as_directory() as checkpoint_dir: data_path = Path(checkpoint_dir) / \"data.pkl\" with open(data_path, \"rb\") as fp: checkpoint_state = pickle.load(fp) start_epoch = checkpoint_state[\"epoch\"] net.load_state_dict(checkpoint_state[\"net_state_dict\"]) optimizer.load_state_dict(checkpoint_state[\"optimizer_state_dict\"]) else: start_epoch = 0 trainset, testset = load_data(data_dir) test_abs = int(len(trainset) * 0.8) train_subset, val_subset = random_split( trainset, [test_abs, len(trainset) - test_abs] ) trainloader = torch.utils.data.DataLoader( train_subset, batch_size=int(config[\"batch_size\"]), shuffle=True, num_workers=8 ) valloader = torch.utils.data.DataLoader( val_subset, batch_size=int(config[\"batch_size\"]), shuffle=True, num_workers=8 ) for epoch in range(start_epoch, 10): # loop over the dataset multiple times running_loss = 0.0 epoch_steps = 0 for i, data in enumerate(trainloader, 0): # get the inputs; data is a list of [inputs, labels] inputs, labels = data inputs, labels = inputs.to(device), labels.to(device) # zero the parameter gradients optimizer.zero_grad() # forward + backward + optimize outputs = net(inputs) loss = criterion(outputs, labels) loss.backward() optimizer.step() # print statistics running_loss += loss.item() epoch_steps += 1 if i % 2000 == 1999: # print every 2000 mini-batches print( \"[%d, %5d] loss: %.3f\" % (epoch + 1, i + 1, running_loss / epoch_steps) ) running_loss = 0.0 # Validation loss val_loss = 0.0 val_steps = 0 total = 0 correct = 0 for i, data in enumerate(valloader, 0): with torch.no_grad(): inputs, labels = data inputs, labels = inputs.to(device), labels.to(device) outputs = net(inputs) _, predicted = torch.max(outputs.data, 1) total += labels.size(0) correct += (predicted == labels).sum().item() loss = criterion(outputs, labels) val_loss += loss.cpu().numpy() val_steps += 1 checkpoint_data = { \"epoch\": epoch, \"net_state_dict\": net.state_dict(), \"optimizer_state_dict\": optimizer.state_dict(), } with tempfile.TemporaryDirectory() as checkpoint_dir: data_path = Path(checkpoint_dir) / \"data.pkl\" with open(data_path, \"wb\") as fp: pickle.dump(checkpoint_data, fp) checkpoint = Checkpoint.from_directory(checkpoint_dir) train.report( {\"loss\": val_loss / val_steps, \"accuracy\": correct / total}, checkpoint=checkpoint, ) print(\"Finished Training\") \ubcf4\ub2e4\uc2dc\ud53c, \ub300\ubd80\ubd84\uc758 \ucf54\ub4dc\ub294 \uc6d0\ubcf8 \uc608\uc81c\uc5d0\uc11c \uc9c1\uc811 \uc801\uc6a9\ub418\uc5c8\uc2b5\ub2c8\ub2e4. \ud14c\uc2a4\ud2b8\uc14b \uc815\ud655\ub3c4(Test set accuracy)# \uc77c\ubc18\uc801\uc73c\ub85c \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc758 \uc131\ub2a5\uc740 \ubaa8\ub378 \ud559\uc2b5 \uc2dc \uc0ac\uc6a9\ud558\uc9c0 \uc54a\uc740 \ub370\uc774\ud130\ub97c \ud14c\uc2a4\ud2b8\uc14b\uc73c\ub85c \ub530\ub85c \ub5bc\uc5b4\ub0b8 \ub4a4, \uc774\ub97c \uc0ac\uc6a9\ud558\uc5ec \ud14c\uc2a4\ud2b8\ud569\ub2c8\ub2e4. \uc774\ub7ec\ud55c \ud14c\uc2a4\ud2b8\uc14b \ub610\ud55c \ud568\uc218\ub85c \uac10\uc2f8\ub458 \uc218 \uc788\uc2b5\ub2c8\ub2e4: def test_accuracy(net, device=\"cpu\"): trainset, testset = load_data() testloader = torch.utils.data.DataLoader( testset, batch_size=4, shuffle=False, num_workers=2 ) correct = 0 total = 0 with torch.no_grad(): for data in testloader: images, labels = data images, labels = images.to(device), labels.to(device) outputs = net(images) _, predicted = torch.max(outputs.data, 1) total += labels.size(0) correct += (predicted == labels).sum().item() return correct / total \uc774 \ud568\uc218\ub294 \ub610\ud55c device \ud30c\ub77c\ubbf8\ud130\ub97c \uc694\uad6c\ud558\ubbc0\ub85c, test set \ud3c9\uac00\ub97c GPU\uc5d0\uc11c \uc218\ud589\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uac80\uc0c9 \uacf5\uac04 \uad6c\uc131# \ub9c8\uc9c0\ub9c9\uc73c\ub85c Ray Tune\uc758 \uac80\uc0c9 \uacf5\uac04\uc744 \uc815\uc758\ud574\uc57c \ud569\ub2c8\ub2e4. \uc608\uc2dc\ub294 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4: config = { \"l1\": tune.choice([2 ** i for i in range(9)]), \"l2\": tune.choice([2 ** i for i in range(9)]), \"lr\": tune.loguniform(1e-4, 1e-1), \"batch_size\": tune.choice([2, 4, 8, 16]) } tune.choice() \ud568\uc218\ub294 \uade0\uc77c\ud558\uac8c \uc0d8\ud50c\ub9c1\ub41c \uac12\ub4e4\uc758 \ubaa9\ub85d\uc744 \uc785\ub825\uc73c\ub85c \ubc1b\uc2b5\ub2c8\ub2e4. \uc704 \uc608\uc2dc\uc5d0\uc11c l1 \ubc0f l2 \ud30c\ub77c\ubbf8\ud130\ub294 4\uc640 256 \uc0ac\uc774\uc758 2\uc758 \uac70\ub4ed\uc81c\uacf1 \uac12\uc778 4, 8, 16, 32, 64, 128, 256 \uc785\ub2c8\ub2e4. lr (\ud559\uc2b5\ub960)\uc740 0.0001\uacfc 0.1 \uc0ac\uc774\uc5d0\uc11c \uade0\uc77c\ud558\uac8c \uc0d8\ud50c\ub9c1 \ub418\uc5b4\uc57c \ud569\ub2c8\ub2e4. \ub9c8\uc9c0\ub9c9\uc73c\ub85c, \ubc30\uce58 \ud06c\uae30\ub294 2, 4, 8, 16\uc911\uc5d0\uc11c \uc120\ud0dd\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uac01 \uc2e4\ud5d8\uc5d0\uc11c, Ray Tune\uc740 \uc774\uc81c \uc774\ub7ec\ud55c \uac80\uc0c9 \uacf5\uac04\uc5d0\uc11c \ub9e4\uac1c\ubcc0\uc218 \uc870\ud569\uc744 \ubb34\uc791\uc704\ub85c \uc0d8\ud50c\ub9c1\ud569\ub2c8\ub2e4. \uadf8\ub7f0 \ub2e4\uc74c \uc5ec\ub7ec \ubaa8\ub378\uc744 \ubcd1\ub82c\ub85c \ud6c8\ub828\ud558\uace0 \uc774 \uc911\uc5d0\uc11c \uac00\uc7a5 \uc131\ub2a5\uc774 \uc88b\uc740 \ubaa8\ub378\uc744 \ucc3e\uc2b5\ub2c8\ub2e4. \ub610\ud55c \uc131\ub2a5\uc774 \uc88b\uc9c0 \uc54a\uc740 \uc2e4\ud5d8\uc744 \uc870\uae30\uc5d0 \uc885\ub8cc\ud558\ub294 ASHAScheduler \ub97c \uc0ac\uc6a9\ud569\ub2c8\ub2e4. \uc0c1\uc218 data_dir \ud30c\ub77c\ubbf8\ud130\ub97c \uc124\uc815\ud558\uae30 \uc704\ud574 functools.partial \ub85c train_cifar \ud568\uc218\ub97c \uac10\uc2f8\ub461\ub2c8\ub2e4. \ub610\ud55c \uac01 \uc2e4\ud5d8\uc5d0 \uc0ac\uc6a9\ud560 \uc218 \uc788\ub294 \uc790\uc6d0\ub4e4(resources)\uc744 Ray Tune\uc5d0 \uc54c\ub9b4 \uc218 \uc788\uc2b5\ub2c8\ub2e4. gpus_per_trial = 2 # ... result = tune.run( partial(train_cifar, data_dir=data_dir), resources_per_trial={\"cpu\": 8, \"gpu\": gpus_per_trial}, config=config, num_samples=num_samples, scheduler=scheduler, checkpoint_at_end=True) \ud30c\uc774\ud1a0\uce58 DataLoader \uc778\uc2a4\ud134\uc2a4\uc758 num_workers \uc744 \ub298\ub9ac\uae30 \uc704\ud574 CPU \uc218\ub97c \uc9c0\uc815\ud558\uace0 \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uac01 \uc2e4\ud5d8\uc5d0\uc11c \uc120\ud0dd\ud55c \uc218\uc758 GPU\ub4e4\uc740 \ud30c\uc774\ud1a0\uce58\uc5d0 \ud45c\uc2dc\ub429\ub2c8\ub2e4. \uc2e4\ud5d8\ub4e4\uc740 \uc694\uccad\ub418\uc9c0 \uc54a\uc740 GPU\uc5d0 \uc561\uc138\uc2a4\ud560 \uc218 \uc5c6\uc73c\ubbc0\ub85c \uac19\uc740 \uc790\uc6d0\ub4e4\uc744 \uc0ac\uc6a9\ud558\ub294 \uc911\ubcf5\ub41c \uc2e4\ud5d8\uc5d0 \ub300\ud574 \uc2e0\uacbd\uc4f0\uc9c0 \uc54a\uc544\ub3c4 \ub429\ub2c8\ub2e4. \ubd80\ubd84 GPUs\ub97c \uc9c0\uc815\ud560 \uc218\ub3c4 \uc788\uc73c\ubbc0\ub85c, gpus_per_trial=0.5 \uc640 \uac19\uc740 \uac83 \ub610\ud55c \uac00\ub2a5\ud569\ub2c8\ub2e4. \uc774\ud6c4 \uac01 \uc2e4\ud5d8\uc740 GPU\ub97c \uacf5\uc720\ud569\ub2c8\ub2e4. \uc0ac\uc6a9\uc790\ub294 \ubaa8\ub378\uc774 \uc5ec\uc804\ud788 GPU\uba54\ubaa8\ub9ac\uc5d0 \uc801\ud569\ud55c\uc9c0\ub9cc \ud655\uc778\ud558\uba74 \ub429\ub2c8\ub2e4. \ubaa8\ub378\uc744 \ud6c8\ub828\uc2dc\ud0a8 \ud6c4, \uac00\uc7a5 \uc131\ub2a5\uc774 \uc88b\uc740 \ubaa8\ub378\uc744 \ucc3e\uace0 \uccb4\ud06c\ud3ec\uc778\ud2b8 \ud30c\uc77c\uc5d0\uc11c \ud559\uc2b5\ub41c \ubaa8\ub378\uc744 \ub85c\ub4dc\ud569\ub2c8\ub2e4. \uc774\ud6c4 test set \uc815\ud655\ub3c4(accuracy)\ub97c \uc5bb\uace0 \ubaa8\ub4e0 \uac83\ub4e4\uc744 \ucd9c\ub825\ud558\uc5ec \ud655\uc778\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc804\uccb4 \uc8fc\uc694 \uae30\ub2a5\uc740 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4. def main(num_samples=10, max_num_epochs=10, gpus_per_trial=2): data_dir = os.path.abspath(\"./data\") load_data(data_dir) config = { \"l1\": tune.choice([2**i for i in range(9)]), \"l2\": tune.choice([2**i for i in range(9)]), \"lr\": tune.loguniform(1e-4, 1e-1), \"batch_size\": tune.choice([2, 4, 8, 16]), } scheduler = ASHAScheduler( metric=\"loss\", mode=\"min\", max_t=max_num_epochs, grace_period=1, reduction_factor=2, ) result = tune.run( partial(train_cifar, data_dir=data_dir), resources_per_trial={\"cpu\": 2, \"gpu\": gpus_per_trial}, config=config, num_samples=num_samples, scheduler=scheduler, ) best_trial = result.get_best_trial(\"loss\", \"min\", \"last\") print(f\"Best trial config: {best_trial.config}\") print(f\"Best trial final validation loss: {best_trial.last_result[\u0027loss\u0027]}\") print(f\"Best trial final validation accuracy: {best_trial.last_result[\u0027accuracy\u0027]}\") best_trained_model = Net(best_trial.config[\"l1\"], best_trial.config[\"l2\"]) device = \"cpu\" if torch.cuda.is_available(): device = \"cuda:0\" if gpus_per_trial \u003e 1: best_trained_model = nn.DataParallel(best_trained_model) best_trained_model.to(device) best_checkpoint = result.get_best_checkpoint(trial=best_trial, metric=\"accuracy\", mode=\"max\") with best_checkpoint.as_directory() as checkpoint_dir: data_path = Path(checkpoint_dir) / \"data.pkl\" with open(data_path, \"rb\") as fp: best_checkpoint_data = pickle.load(fp) best_trained_model.load_state_dict(best_checkpoint_data[\"net_state_dict\"]) test_acc = test_accuracy(best_trained_model, device) print(\"Best trial test set accuracy: {}\".format(test_acc)) if __name__ == \"__main__\": # \ub9e4 \uc2e4\ud5d8\ub2f9 \uc0ac\uc6a9\ud560 GPU \uc218\ub97c \uc5ec\uae30\uc5d0\uc11c \ubcc0\uacbd\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4: main(num_samples=10, max_num_epochs=10, gpus_per_trial=0) 0%| | 0.00/170M [00:00\u003c?, ?B/s] 0%| | 32.8k/170M [00:00\u003c14:35, 195kB/s] 0%| | 65.5k/170M [00:00\u003c14:38, 194kB/s] 0%| | 98.3k/170M [00:00\u003c14:21, 198kB/s] 0%| | 229k/170M [00:00\u003c06:36, 430kB/s] 0%| | 459k/170M [00:00\u003c03:40, 772kB/s] 1%| | 918k/170M [00:01\u003c01:57, 1.44MB/s] 1%| | 1.84M/170M [00:01\u003c01:01, 2.76MB/s] 2%|\u258f | 3.70M/170M [00:01\u003c00:30, 5.43MB/s] 4%|\u258d | 7.37M/170M [00:01\u003c00:15, 10.6MB/s] 6%|\u258c | 10.4M/170M [00:01\u003c00:13, 12.2MB/s] 8%|\u258a | 13.3M/170M [00:01\u003c00:11, 13.8MB/s] 10%|\u2589 | 16.3M/170M [00:02\u003c00:10, 14.8MB/s] 11%|\u2588 | 19.1M/170M [00:02\u003c00:09, 15.4MB/s] 13%|\u2588\u258e | 22.1M/170M [00:02\u003c00:09, 15.9MB/s] 15%|\u2588\u258d | 25.1M/170M [00:02\u003c00:08, 16.4MB/s] 16%|\u2588\u258b | 28.1M/170M [00:02\u003c00:08, 16.7MB/s] 18%|\u2588\u258a | 30.9M/170M [00:02\u003c00:08, 16.8MB/s] 20%|\u2588\u2589 | 33.8M/170M [00:03\u003c00:08, 16.8MB/s] 22%|\u2588\u2588\u258f | 36.8M/170M [00:03\u003c00:07, 16.9MB/s] 23%|\u2588\u2588\u258e | 39.6M/170M [00:03\u003c00:07, 16.9MB/s] 25%|\u2588\u2588\u258c | 42.7M/170M [00:03\u003c00:07, 17.1MB/s] 27%|\u2588\u2588\u258b | 45.6M/170M [00:03\u003c00:07, 17.1MB/s] 28%|\u2588\u2588\u258a | 48.5M/170M [00:03\u003c00:07, 17.1MB/s] 30%|\u2588\u2588\u2588 | 51.5M/170M [00:04\u003c00:06, 17.1MB/s] 32%|\u2588\u2588\u2588\u258f | 54.3M/170M [00:04\u003c00:06, 17.0MB/s] 34%|\u2588\u2588\u2588\u258e | 57.3M/170M [00:04\u003c00:06, 17.2MB/s] 35%|\u2588\u2588\u2588\u258c | 60.3M/170M [00:04\u003c00:06, 17.2MB/s] 37%|\u2588\u2588\u2588\u258b | 63.2M/170M [00:04\u003c00:06, 17.1MB/s] 39%|\u2588\u2588\u2588\u2589 | 66.2M/170M [00:04\u003c00:06, 17.2MB/s] 41%|\u2588\u2588\u2588\u2588 | 69.1M/170M [00:05\u003c00:05, 17.1MB/s] 42%|\u2588\u2588\u2588\u2588\u258f | 72.1M/170M [00:05\u003c00:05, 17.1MB/s] 44%|\u2588\u2588\u2588\u2588\u258d | 75.0M/170M [00:05\u003c00:05, 17.1MB/s] 46%|\u2588\u2588\u2588\u2588\u258c | 78.0M/170M [00:05\u003c00:05, 17.2MB/s] 47%|\u2588\u2588\u2588\u2588\u258b | 80.8M/170M [00:05\u003c00:05, 17.1MB/s] 49%|\u2588\u2588\u2588\u2588\u2589 | 83.8M/170M [00:05\u003c00:05, 17.1MB/s] 51%|\u2588\u2588\u2588\u2588\u2588 | 86.8M/170M [00:06\u003c00:04, 17.1MB/s] 53%|\u2588\u2588\u2588\u2588\u2588\u258e | 89.6M/170M [00:06\u003c00:04, 17.1MB/s] 54%|\u2588\u2588\u2588\u2588\u2588\u258d | 92.6M/170M [00:06\u003c00:04, 17.2MB/s] 56%|\u2588\u2588\u2588\u2588\u2588\u258c | 95.5M/170M [00:06\u003c00:04, 17.1MB/s] 58%|\u2588\u2588\u2588\u2588\u2588\u258a | 98.5M/170M [00:06\u003c00:04, 17.2MB/s] 60%|\u2588\u2588\u2588\u2588\u2588\u2589 | 101M/170M [00:07\u003c00:04, 17.2MB/s] 61%|\u2588\u2588\u2588\u2588\u2588\u2588 | 104M/170M [00:07\u003c00:03, 17.1MB/s] 63%|\u2588\u2588\u2588\u2588\u2588\u2588\u258e | 107M/170M [00:07\u003c00:03, 17.1MB/s] 65%|\u2588\u2588\u2588\u2588\u2588\u2588\u258d | 110M/170M [00:07\u003c00:03, 17.2MB/s] 66%|\u2588\u2588\u2588\u2588\u2588\u2588\u258b | 113M/170M [00:07\u003c00:03, 17.2MB/s] 68%|\u2588\u2588\u2588\u2588\u2588\u2588\u258a | 116M/170M [00:07\u003c00:03, 17.2MB/s] 70%|\u2588\u2588\u2588\u2588\u2588\u2588\u2589 | 119M/170M [00:08\u003c00:02, 17.2MB/s] 72%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258f | 122M/170M [00:08\u003c00:02, 17.3MB/s] 73%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258e | 125M/170M [00:08\u003c00:02, 17.3MB/s] 75%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258c | 128M/170M [00:08\u003c00:02, 17.2MB/s] 77%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258b | 131M/170M [00:08\u003c00:02, 17.2MB/s] 79%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258a | 134M/170M [00:08\u003c00:02, 17.2MB/s] 80%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588 | 137M/170M [00:09\u003c00:01, 17.2MB/s] 82%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258f | 140M/170M [00:09\u003c00:01, 17.1MB/s] 84%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258e | 143M/170M [00:09\u003c00:01, 17.0MB/s] 85%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258c | 146M/170M [00:09\u003c00:01, 17.1MB/s] 87%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258b | 149M/170M [00:09\u003c00:01, 17.0MB/s] 89%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2589 | 152M/170M [00:09\u003c00:01, 16.9MB/s] 91%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588 | 155M/170M [00:10\u003c00:00, 17.0MB/s] 92%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258f| 157M/170M [00:10\u003c00:00, 17.1MB/s] 94%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258d| 160M/170M [00:10\u003c00:00, 17.0MB/s] 96%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258c| 163M/170M [00:10\u003c00:00, 17.1MB/s] 97%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u258b| 166M/170M [00:10\u003c00:00, 17.1MB/s] 99%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2589| 169M/170M [00:10\u003c00:00, 17.2MB/s] 100%|\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588\u2588| 170M/170M [00:10\u003c00:00, 15.5MB/s] 2025-10-03 22:26:01,958 INFO worker.py:1642 -- Started a local Ray instance. 2025-10-03 22:26:04,049 INFO tune.py:228 -- Initializing Ray automatically. For cluster usage or custom Ray initialization, call `ray.init(...)` before `tune.run(...)`. 2025-10-03 22:26:04,051 INFO tune.py:654 -- [output] This will use the new output engine with verbosity 2. To disable the new output and use the legacy output engine, set the environment variable RAY_AIR_NEW_OUTPUT=0. For more information, please see https://github.com/ray-project/ray/issues/36949 \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Configuration for experiment train_cifar_2025-10-03_22-26-04 \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 Search algorithm BasicVariantGenerator \u2502 \u2502 Scheduler AsyncHyperBandScheduler \u2502 \u2502 Number of trials 10 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f View detailed results here: /root/ray_results/train_cifar_2025-10-03_22-26-04 To visualize your results with TensorBoard, run: `tensorboard --logdir /root/ray_results/train_cifar_2025-10-03_22-26-04` Trial status: 10 PENDING Current time: 2025-10-03 22:26:04. Total running time: 0s Logical resource usage: 20.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00000 PENDING 2 256 0.000305994 16 \u2502 \u2502 train_cifar_f2b95_00001 PENDING 128 32 0.010341 16 \u2502 \u2502 train_cifar_f2b95_00002 PENDING 4 256 0.000582548 4 \u2502 \u2502 train_cifar_f2b95_00003 PENDING 8 2 0.03878 4 \u2502 \u2502 train_cifar_f2b95_00004 PENDING 128 64 0.0275418 2 \u2502 \u2502 train_cifar_f2b95_00005 PENDING 4 8 0.000769138 4 \u2502 \u2502 train_cifar_f2b95_00006 PENDING 64 8 0.00236933 16 \u2502 \u2502 train_cifar_f2b95_00007 PENDING 8 128 0.00365739 2 \u2502 \u2502 train_cifar_f2b95_00008 PENDING 16 128 0.000192995 16 \u2502 \u2502 train_cifar_f2b95_00009 PENDING 8 256 0.00117126 8 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources (raylet) /opt/conda/lib/python3.11/site-packages/ray/_private/parameter.py:4: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools\u003c81. (raylet) import pkg_resources Trial train_cifar_f2b95_00009 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 8 \u2502 \u2502 l1 8 \u2502 \u2502 l2 256 \u2502 \u2502 lr 0.00117 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 16 \u2502 \u2502 l1 128 \u2502 \u2502 l2 32 \u2502 \u2502 lr 0.01034 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00007 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00007 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 2 \u2502 \u2502 l1 8 \u2502 \u2502 l2 128 \u2502 \u2502 lr 0.00366 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00004 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00004 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 2 \u2502 \u2502 l1 128 \u2502 \u2502 l2 64 \u2502 \u2502 lr 0.02754 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00008 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00008 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 16 \u2502 \u2502 l1 16 \u2502 \u2502 l2 128 \u2502 \u2502 lr 0.00019 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00003 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00003 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 4 \u2502 \u2502 l1 8 \u2502 \u2502 l2 2 \u2502 \u2502 lr 0.03878 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00000 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00000 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 16 \u2502 \u2502 l1 2 \u2502 \u2502 l2 256 \u2502 \u2502 lr 0.00031 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00002 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00002 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 4 \u2502 \u2502 l1 4 \u2502 \u2502 l2 256 \u2502 \u2502 lr 0.00058 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00006 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00006 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 16 \u2502 \u2502 l1 64 \u2502 \u2502 l2 8 \u2502 \u2502 lr 0.00237 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00005 started with configuration: \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00005 config \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 batch_size 4 \u2502 \u2502 l1 4 \u2502 \u2502 l2 8 \u2502 \u2502 lr 0.00077 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12600) [1, 2000] loss: 2.083 Trial train_cifar_f2b95_00001 finished iteration 1 at 2025-10-03 22:26:20. Total running time: 16s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 13.00158 \u2502 \u2502 time_total_s 13.00158 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.4763 \u2502 \u2502 loss 1.45586 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000000) Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00008 finished iteration 1 at 2025-10-03 22:26:21. Total running time: 17s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00008 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 13.74059 \u2502 \u2502 time_total_s 13.74059 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.1092 \u2502 \u2502 loss 2.3008 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00008 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00008_8_batch_size=16,l1=16,l2=128,lr=0.0002_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00008 completed after 1 iterations at 2025-10-03 22:26:21. Total running time: 17s Trial train_cifar_f2b95_00000 finished iteration 1 at 2025-10-03 22:26:21. Total running time: 17s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00000 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 14.05161 \u2502 \u2502 time_total_s 14.05161 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.2076 \u2502 \u2502 loss 2.05941 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00000 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00000_0_batch_size=16,l1=2,l2=256,lr=0.0003_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00000 completed after 1 iterations at 2025-10-03 22:26:21. Total running time: 17s Trial train_cifar_f2b95_00006 finished iteration 1 at 2025-10-03 22:26:22. Total running time: 17s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00006 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 14.296 \u2502 \u2502 time_total_s 14.296 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.3684 \u2502 \u2502 loss 1.73915 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00006 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00006_6_batch_size=16,l1=64,l2=8,lr=0.0024_2025-10-03_22-26-04/checkpoint_000000 (func pid=12594) [1, 4000] loss: 1.162 [repeated 11x across cluster] (Ray deduplicates logs by default. Set RAY_DEDUP_LOGS=0 to disable log deduplication, or see https://docs.ray.io/en/master/ray-observability/ray-logging.html#log-deduplication for more options.) Trial train_cifar_f2b95_00009 finished iteration 1 at 2025-10-03 22:26:27. Total running time: 22s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 19.5243 \u2502 \u2502 time_total_s 19.5243 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.4436 \u2502 \u2502 loss 1.53569 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000000 (func pid=12593) [1, 6000] loss: 0.613 [repeated 7x across cluster] (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000000) [repeated 4x across cluster] Trial train_cifar_f2b95_00001 finished iteration 2 at 2025-10-03 22:26:30. Total running time: 25s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000001 \u2502 \u2502 time_this_iter_s 9.42768 \u2502 \u2502 time_total_s 22.42926 \u2502 \u2502 training_iteration 2 \u2502 \u2502 accuracy 0.5172 \u2502 \u2502 loss 1.35872 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 2 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000001 Trial train_cifar_f2b95_00006 finished iteration 2 at 2025-10-03 22:26:31. Total running time: 27s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00006 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000001 \u2502 \u2502 time_this_iter_s 9.2087 \u2502 \u2502 time_total_s 23.5047 \u2502 \u2502 training_iteration 2 \u2502 \u2502 accuracy 0.4657 \u2502 \u2502 loss 1.48145 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00006 saved a checkpoint for iteration 2 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00006_6_batch_size=16,l1=64,l2=8,lr=0.0024_2025-10-03_22-26-04/checkpoint_000001 Trial train_cifar_f2b95_00006 completed after 2 iterations at 2025-10-03 22:26:31. Total running time: 27s (func pid=12593) [1, 8000] loss: 0.445 [repeated 7x across cluster] Trial status: 3 TERMINATED | 7 RUNNING Current time: 2025-10-03 22:26:34. Total running time: 30s Logical resource usage: 14.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size iter total time (s) loss accuracy \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00001 RUNNING 128 32 0.010341 16 2 22.4293 1.35872 0.5172 \u2502 \u2502 train_cifar_f2b95_00002 RUNNING 4 256 0.000582548 4 \u2502 \u2502 train_cifar_f2b95_00003 RUNNING 8 2 0.03878 4 \u2502 \u2502 train_cifar_f2b95_00004 RUNNING 128 64 0.0275418 2 \u2502 \u2502 train_cifar_f2b95_00005 RUNNING 4 8 0.000769138 4 \u2502 \u2502 train_cifar_f2b95_00007 RUNNING 8 128 0.00365739 2 \u2502 \u2502 train_cifar_f2b95_00009 RUNNING 8 256 0.00117126 8 1 19.5243 1.53569 0.4436 \u2502 \u2502 train_cifar_f2b95_00000 TERMINATED 2 256 0.000305994 16 1 14.0516 2.05941 0.2076 \u2502 \u2502 train_cifar_f2b95_00006 TERMINATED 64 8 0.00236933 16 2 23.5047 1.48145 0.4657 \u2502 \u2502 train_cifar_f2b95_00008 TERMINATED 16 128 0.000192995 16 1 13.7406 2.3008 0.1092 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12593) [1, 10000] loss: 0.339 [repeated 7x across cluster] Trial train_cifar_f2b95_00001 finished iteration 3 at 2025-10-03 22:26:39. Total running time: 35s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000002 \u2502 \u2502 time_this_iter_s 9.45501 \u2502 \u2502 time_total_s 31.88427 \u2502 \u2502 training_iteration 3 \u2502 \u2502 accuracy 0.5172 \u2502 \u2502 loss 1.34038 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 3 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000002 (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000002) [repeated 3x across cluster] Trial train_cifar_f2b95_00003 finished iteration 1 at 2025-10-03 22:26:40. Total running time: 36s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00003 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 32.93671 \u2502 \u2502 time_total_s 32.93671 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.0976 \u2502 \u2502 loss 2.34242 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00003 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00003_3_batch_size=4,l1=8,l2=2,lr=0.0388_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00003 completed after 1 iterations at 2025-10-03 22:26:40. Total running time: 36s Trial train_cifar_f2b95_00002 finished iteration 1 at 2025-10-03 22:26:40. Total running time: 36s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00002 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 33.11155 \u2502 \u2502 time_total_s 33.11155 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.3566 \u2502 \u2502 loss 1.63892 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00002 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00002_2_batch_size=4,l1=4,l2=256,lr=0.0006_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00005 finished iteration 1 at 2025-10-03 22:26:41. Total running time: 37s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00005 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 33.66846 \u2502 \u2502 time_total_s 33.66846 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.3463 \u2502 \u2502 loss 1.70408 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00005 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00005_5_batch_size=4,l1=4,l2=8,lr=0.0008_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00009 finished iteration 2 at 2025-10-03 22:26:43. Total running time: 39s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000001 \u2502 \u2502 time_this_iter_s 16.19927 \u2502 \u2502 time_total_s 35.72356 \u2502 \u2502 training_iteration 2 \u2502 \u2502 accuracy 0.5049 \u2502 \u2502 loss 1.36047 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 2 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000001 (func pid=12598) [1, 14000] loss: 0.276 [repeated 5x across cluster] Trial train_cifar_f2b95_00001 finished iteration 4 at 2025-10-03 22:26:48. Total running time: 44s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000003 \u2502 \u2502 time_this_iter_s 9.31968 \u2502 \u2502 time_total_s 41.20394 \u2502 \u2502 training_iteration 4 \u2502 \u2502 accuracy 0.5287 \u2502 \u2502 loss 1.35828 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000003) [repeated 5x across cluster] Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 4 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000003 (func pid=12596) [1, 14000] loss: 0.334 [repeated 7x across cluster] (func pid=12593) [2, 6000] loss: 0.534 [repeated 6x across cluster] Trial train_cifar_f2b95_00001 finished iteration 5 at 2025-10-03 22:26:57. Total running time: 53s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000004 \u2502 \u2502 time_this_iter_s 9.13369 \u2502 \u2502 time_total_s 50.33764 \u2502 \u2502 training_iteration 5 \u2502 \u2502 accuracy 0.5405 \u2502 \u2502 loss 1.33497 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 5 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000004 (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000004) Trial train_cifar_f2b95_00009 finished iteration 3 at 2025-10-03 22:26:59. Total running time: 54s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000002 \u2502 \u2502 time_this_iter_s 15.66666 \u2502 \u2502 time_total_s 51.39022 \u2502 \u2502 training_iteration 3 \u2502 \u2502 accuracy 0.5157 \u2502 \u2502 loss 1.30244 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 3 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000002 (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000002) (func pid=12595) [2, 8000] loss: 0.401 [repeated 5x across cluster] Trial status: 4 TERMINATED | 6 RUNNING Current time: 2025-10-03 22:27:04. Total running time: 1min 0s Logical resource usage: 12.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size iter total time (s) loss accuracy \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00001 RUNNING 128 32 0.010341 16 5 50.3376 1.33497 0.5405 \u2502 \u2502 train_cifar_f2b95_00002 RUNNING 4 256 0.000582548 4 1 33.1116 1.63892 0.3566 \u2502 \u2502 train_cifar_f2b95_00004 RUNNING 128 64 0.0275418 2 \u2502 \u2502 train_cifar_f2b95_00005 RUNNING 4 8 0.000769138 4 1 33.6685 1.70408 0.3463 \u2502 \u2502 train_cifar_f2b95_00007 RUNNING 8 128 0.00365739 2 \u2502 \u2502 train_cifar_f2b95_00009 RUNNING 8 256 0.00117126 8 3 51.3902 1.30244 0.5157 \u2502 \u2502 train_cifar_f2b95_00000 TERMINATED 2 256 0.000305994 16 1 14.0516 2.05941 0.2076 \u2502 \u2502 train_cifar_f2b95_00003 TERMINATED 8 2 0.03878 4 1 32.9367 2.34242 0.0976 \u2502 \u2502 train_cifar_f2b95_00006 TERMINATED 64 8 0.00236933 16 2 23.5047 1.48145 0.4657 \u2502 \u2502 train_cifar_f2b95_00008 TERMINATED 16 128 0.000192995 16 1 13.7406 2.3008 0.1092 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00007 finished iteration 1 at 2025-10-03 22:27:05. Total running time: 1min 0s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00007 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 57.44913 \u2502 \u2502 time_total_s 57.44913 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.2941 \u2502 \u2502 loss 1.85399 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00007 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00007_7_batch_size=2,l1=8,l2=128,lr=0.0037_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00007 completed after 1 iterations at 2025-10-03 22:27:05. Total running time: 1min 0s (func pid=12598) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00007_7_batch_size=2,l1=8,l2=128,lr=0.0037_2025-10-03_22-26-04/checkpoint_000000) Trial train_cifar_f2b95_00001 finished iteration 6 at 2025-10-03 22:27:07. Total running time: 1min 3s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000005 \u2502 \u2502 time_this_iter_s 9.20401 \u2502 \u2502 time_total_s 59.54165 \u2502 \u2502 training_iteration 6 \u2502 \u2502 accuracy 0.5458 \u2502 \u2502 loss 1.38728 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 6 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000005 (func pid=12596) [1, 20000] loss: 0.233 [repeated 6x across cluster] (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000005) Trial train_cifar_f2b95_00002 finished iteration 2 at 2025-10-03 22:27:09. Total running time: 1min 5s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00002 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000001 \u2502 \u2502 time_this_iter_s 28.47496 \u2502 \u2502 time_total_s 61.58651 \u2502 \u2502 training_iteration 2 \u2502 \u2502 accuracy 0.4365 \u2502 \u2502 loss 1.50182 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00002 saved a checkpoint for iteration 2 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00002_2_batch_size=4,l1=4,l2=256,lr=0.0006_2025-10-03_22-26-04/checkpoint_000001 Trial train_cifar_f2b95_00002 completed after 2 iterations at 2025-10-03 22:27:09. Total running time: 1min 5s Trial train_cifar_f2b95_00005 finished iteration 2 at 2025-10-03 22:27:09. Total running time: 1min 5s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00005 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000001 \u2502 \u2502 time_this_iter_s 28.50694 \u2502 \u2502 time_total_s 62.1754 \u2502 \u2502 training_iteration 2 \u2502 \u2502 accuracy 0.4021 \u2502 \u2502 loss 1.57171 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00005 saved a checkpoint for iteration 2 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00005_5_batch_size=4,l1=4,l2=8,lr=0.0008_2025-10-03_22-26-04/checkpoint_000001 Trial train_cifar_f2b95_00005 completed after 2 iterations at 2025-10-03 22:27:09. Total running time: 1min 5s Trial train_cifar_f2b95_00004 finished iteration 1 at 2025-10-03 22:27:13. Total running time: 1min 9s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00004 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000000 \u2502 \u2502 time_this_iter_s 66.0309 \u2502 \u2502 time_total_s 66.0309 \u2502 \u2502 training_iteration 1 \u2502 \u2502 accuracy 0.0981 \u2502 \u2502 loss 2.32891 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00004 saved a checkpoint for iteration 1 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00004_4_batch_size=2,l1=128,l2=64,lr=0.0275_2025-10-03_22-26-04/checkpoint_000000 Trial train_cifar_f2b95_00004 completed after 1 iterations at 2025-10-03 22:27:13. Total running time: 1min 9s (func pid=12596) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00004_4_batch_size=2,l1=128,l2=64,lr=0.0275_2025-10-03_22-26-04/checkpoint_000000) [repeated 3x across cluster] (func pid=12592) [7, 2000] loss: 1.145 [repeated 2x across cluster] Trial train_cifar_f2b95_00009 finished iteration 4 at 2025-10-03 22:27:14. Total running time: 1min 10s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000003 \u2502 \u2502 time_this_iter_s 15.57003 \u2502 \u2502 time_total_s 66.96025 \u2502 \u2502 training_iteration 4 \u2502 \u2502 accuracy 0.5614 \u2502 \u2502 loss 1.22414 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 4 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000003 Trial train_cifar_f2b95_00001 finished iteration 7 at 2025-10-03 22:27:16. Total running time: 1min 12s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000006 \u2502 \u2502 time_this_iter_s 9.23141 \u2502 \u2502 time_total_s 68.77305 \u2502 \u2502 training_iteration 7 \u2502 \u2502 accuracy 0.551 \u2502 \u2502 loss 1.34653 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 7 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000006 (func pid=12600) [5, 2000] loss: 1.201 (func pid=12592) [8, 2000] loss: 1.135 Trial train_cifar_f2b95_00001 finished iteration 8 at 2025-10-03 22:27:25. Total running time: 1min 21s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000007 \u2502 \u2502 time_this_iter_s 9.14384 \u2502 \u2502 time_total_s 77.91689 \u2502 \u2502 training_iteration 8 \u2502 \u2502 accuracy 0.5385 \u2502 \u2502 loss 1.37364 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 8 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000007 (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000007) [repeated 3x across cluster] Trial train_cifar_f2b95_00009 finished iteration 5 at 2025-10-03 22:27:30. Total running time: 1min 26s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000004 \u2502 \u2502 time_this_iter_s 16.20591 \u2502 \u2502 time_total_s 83.16616 \u2502 \u2502 training_iteration 5 \u2502 \u2502 accuracy 0.5578 \u2502 \u2502 loss 1.2334 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 5 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000004 (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000004) (func pid=12592) [9, 2000] loss: 1.111 [repeated 2x across cluster] Trial status: 8 TERMINATED | 2 RUNNING Current time: 2025-10-03 22:27:34. Total running time: 1min 30s Logical resource usage: 4.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size iter total time (s) loss accuracy \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00001 RUNNING 128 32 0.010341 16 8 77.9169 1.37364 0.5385 \u2502 \u2502 train_cifar_f2b95_00009 RUNNING 8 256 0.00117126 8 5 83.1662 1.2334 0.5578 \u2502 \u2502 train_cifar_f2b95_00000 TERMINATED 2 256 0.000305994 16 1 14.0516 2.05941 0.2076 \u2502 \u2502 train_cifar_f2b95_00002 TERMINATED 4 256 0.000582548 4 2 61.5865 1.50182 0.4365 \u2502 \u2502 train_cifar_f2b95_00003 TERMINATED 8 2 0.03878 4 1 32.9367 2.34242 0.0976 \u2502 \u2502 train_cifar_f2b95_00004 TERMINATED 128 64 0.0275418 2 1 66.0309 2.32891 0.0981 \u2502 \u2502 train_cifar_f2b95_00005 TERMINATED 4 8 0.000769138 4 2 62.1754 1.57171 0.4021 \u2502 \u2502 train_cifar_f2b95_00006 TERMINATED 64 8 0.00236933 16 2 23.5047 1.48145 0.4657 \u2502 \u2502 train_cifar_f2b95_00007 TERMINATED 8 128 0.00365739 2 1 57.4491 1.85399 0.2941 \u2502 \u2502 train_cifar_f2b95_00008 TERMINATED 16 128 0.000192995 16 1 13.7406 2.3008 0.1092 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 finished iteration 9 at 2025-10-03 22:27:35. Total running time: 1min 31s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000008 \u2502 \u2502 time_this_iter_s 9.95506 \u2502 \u2502 time_total_s 87.87195 \u2502 \u2502 training_iteration 9 \u2502 \u2502 accuracy 0.5591 \u2502 \u2502 loss 1.31092 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 9 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000008 (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000008) (func pid=12600) [6, 4000] loss: 0.585 [repeated 2x across cluster] Trial train_cifar_f2b95_00001 finished iteration 10 at 2025-10-03 22:27:45. Total running time: 1min 41s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00001 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000009 \u2502 \u2502 time_this_iter_s 9.5594 \u2502 \u2502 time_total_s 97.43135 \u2502 \u2502 training_iteration 10 \u2502 \u2502 accuracy 0.5712 \u2502 \u2502 loss 1.29955 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00001 saved a checkpoint for iteration 10 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000009 Trial train_cifar_f2b95_00001 completed after 10 iterations at 2025-10-03 22:27:45. Total running time: 1min 41s (func pid=12592) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00001_1_batch_size=16,l1=128,l2=32,lr=0.0103_2025-10-03_22-26-04/checkpoint_000009) Trial train_cifar_f2b95_00009 finished iteration 6 at 2025-10-03 22:27:46. Total running time: 1min 42s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000005 \u2502 \u2502 time_this_iter_s 16.12082 \u2502 \u2502 time_total_s 99.28699 \u2502 \u2502 training_iteration 6 \u2502 \u2502 accuracy 0.5753 \u2502 \u2502 loss 1.20882 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 6 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000005 (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000005) (func pid=12600) [7, 2000] loss: 1.131 [repeated 2x across cluster] (func pid=12600) [7, 4000] loss: 0.576 Trial train_cifar_f2b95_00009 finished iteration 7 at 2025-10-03 22:28:03. Total running time: 1min 58s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000006 \u2502 \u2502 time_this_iter_s 16.09692 \u2502 \u2502 time_total_s 115.38391 \u2502 \u2502 training_iteration 7 \u2502 \u2502 accuracy 0.5877 \u2502 \u2502 loss 1.15432 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 7 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000006 (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000006) Trial status: 9 TERMINATED | 1 RUNNING Current time: 2025-10-03 22:28:04. Total running time: 2min 0s Logical resource usage: 2.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size iter total time (s) loss accuracy \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00009 RUNNING 8 256 0.00117126 8 7 115.384 1.15432 0.5877 \u2502 \u2502 train_cifar_f2b95_00000 TERMINATED 2 256 0.000305994 16 1 14.0516 2.05941 0.2076 \u2502 \u2502 train_cifar_f2b95_00001 TERMINATED 128 32 0.010341 16 10 97.4314 1.29955 0.5712 \u2502 \u2502 train_cifar_f2b95_00002 TERMINATED 4 256 0.000582548 4 2 61.5865 1.50182 0.4365 \u2502 \u2502 train_cifar_f2b95_00003 TERMINATED 8 2 0.03878 4 1 32.9367 2.34242 0.0976 \u2502 \u2502 train_cifar_f2b95_00004 TERMINATED 128 64 0.0275418 2 1 66.0309 2.32891 0.0981 \u2502 \u2502 train_cifar_f2b95_00005 TERMINATED 4 8 0.000769138 4 2 62.1754 1.57171 0.4021 \u2502 \u2502 train_cifar_f2b95_00006 TERMINATED 64 8 0.00236933 16 2 23.5047 1.48145 0.4657 \u2502 \u2502 train_cifar_f2b95_00007 TERMINATED 8 128 0.00365739 2 1 57.4491 1.85399 0.2941 \u2502 \u2502 train_cifar_f2b95_00008 TERMINATED 16 128 0.000192995 16 1 13.7406 2.3008 0.1092 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12600) [8, 2000] loss: 1.116 (func pid=12600) [8, 4000] loss: 0.563 Trial train_cifar_f2b95_00009 finished iteration 8 at 2025-10-03 22:28:19. Total running time: 2min 14s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000007 \u2502 \u2502 time_this_iter_s 16.00349 \u2502 \u2502 time_total_s 131.38739 \u2502 \u2502 training_iteration 8 \u2502 \u2502 accuracy 0.5892 \u2502 \u2502 loss 1.172 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 8 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000007 (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000007) (func pid=12600) [9, 2000] loss: 1.089 (func pid=12600) [9, 4000] loss: 0.556 Trial status: 9 TERMINATED | 1 RUNNING Current time: 2025-10-03 22:28:34. Total running time: 2min 30s Logical resource usage: 2.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size iter total time (s) loss accuracy \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00009 RUNNING 8 256 0.00117126 8 8 131.387 1.172 0.5892 \u2502 \u2502 train_cifar_f2b95_00000 TERMINATED 2 256 0.000305994 16 1 14.0516 2.05941 0.2076 \u2502 \u2502 train_cifar_f2b95_00001 TERMINATED 128 32 0.010341 16 10 97.4314 1.29955 0.5712 \u2502 \u2502 train_cifar_f2b95_00002 TERMINATED 4 256 0.000582548 4 2 61.5865 1.50182 0.4365 \u2502 \u2502 train_cifar_f2b95_00003 TERMINATED 8 2 0.03878 4 1 32.9367 2.34242 0.0976 \u2502 \u2502 train_cifar_f2b95_00004 TERMINATED 128 64 0.0275418 2 1 66.0309 2.32891 0.0981 \u2502 \u2502 train_cifar_f2b95_00005 TERMINATED 4 8 0.000769138 4 2 62.1754 1.57171 0.4021 \u2502 \u2502 train_cifar_f2b95_00006 TERMINATED 64 8 0.00236933 16 2 23.5047 1.48145 0.4657 \u2502 \u2502 train_cifar_f2b95_00007 TERMINATED 8 128 0.00365739 2 1 57.4491 1.85399 0.2941 \u2502 \u2502 train_cifar_f2b95_00008 TERMINATED 16 128 0.000192995 16 1 13.7406 2.3008 0.1092 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 finished iteration 9 at 2025-10-03 22:28:34. Total running time: 2min 30s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000008 \u2502 \u2502 time_this_iter_s 15.62216 \u2502 \u2502 time_total_s 147.00955 \u2502 \u2502 training_iteration 9 \u2502 \u2502 accuracy 0.5913 \u2502 \u2502 loss 1.16831 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000008) Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 9 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000008 (func pid=12600) [10, 2000] loss: 1.090 (func pid=12600) [10, 4000] loss: 0.545 Trial train_cifar_f2b95_00009 finished iteration 10 at 2025-10-03 22:28:51. Total running time: 2min 47s \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial train_cifar_f2b95_00009 result \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 checkpoint_dir_name checkpoint_000009 \u2502 \u2502 time_this_iter_s 16.46125 \u2502 \u2502 time_total_s 163.47079 \u2502 \u2502 training_iteration 10 \u2502 \u2502 accuracy 0.6025 \u2502 \u2502 loss 1.1439 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f Trial train_cifar_f2b95_00009 saved a checkpoint for iteration 10 at: (local)/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000009 Trial train_cifar_f2b95_00009 completed after 10 iterations at 2025-10-03 22:28:51. Total running time: 2min 47s Trial status: 10 TERMINATED Current time: 2025-10-03 22:28:51. Total running time: 2min 47s Logical resource usage: 2.0/256 CPUs, 0/8 GPUs (0.0/1.0 accelerator_type:H200) \u256d\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256e \u2502 Trial name status l1 l2 lr batch_size iter total time (s) loss accuracy \u2502 \u251c\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2524 \u2502 train_cifar_f2b95_00000 TERMINATED 2 256 0.000305994 16 1 14.0516 2.05941 0.2076 \u2502 \u2502 train_cifar_f2b95_00001 TERMINATED 128 32 0.010341 16 10 97.4314 1.29955 0.5712 \u2502 \u2502 train_cifar_f2b95_00002 TERMINATED 4 256 0.000582548 4 2 61.5865 1.50182 0.4365 \u2502 \u2502 train_cifar_f2b95_00003 TERMINATED 8 2 0.03878 4 1 32.9367 2.34242 0.0976 \u2502 \u2502 train_cifar_f2b95_00004 TERMINATED 128 64 0.0275418 2 1 66.0309 2.32891 0.0981 \u2502 \u2502 train_cifar_f2b95_00005 TERMINATED 4 8 0.000769138 4 2 62.1754 1.57171 0.4021 \u2502 \u2502 train_cifar_f2b95_00006 TERMINATED 64 8 0.00236933 16 2 23.5047 1.48145 0.4657 \u2502 \u2502 train_cifar_f2b95_00007 TERMINATED 8 128 0.00365739 2 1 57.4491 1.85399 0.2941 \u2502 \u2502 train_cifar_f2b95_00008 TERMINATED 16 128 0.000192995 16 1 13.7406 2.3008 0.1092 \u2502 \u2502 train_cifar_f2b95_00009 TERMINATED 8 256 0.00117126 8 10 163.471 1.1439 0.6025 \u2502 \u2570\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u2500\u256f (func pid=12600) Checkpoint successfully created at: Checkpoint(filesystem=local, path=/root/ray_results/train_cifar_2025-10-03_22-26-04/train_cifar_f2b95_00009_9_batch_size=8,l1=8,l2=256,lr=0.0012_2025-10-03_22-26-04/checkpoint_000009) Best trial config: {\u0027l1\u0027: 8, \u0027l2\u0027: 256, \u0027lr\u0027: 0.001171259491329369, \u0027batch_size\u0027: 8} Best trial final validation loss: 1.143903388774395 Best trial final validation accuracy: 0.6025 Best trial test set accuracy: 0.5994 \ucf54\ub4dc\ub97c \uc2e4\ud589\ud558\uba74 \uacb0\uacfc\ub294 \ub2e4\uc74c\uacfc \uac19\uc774 \ub098\uc62c \uac83\uc785\ub2c8\ub2e4: Number of trials: 10/10 (10 TERMINATED) +-----+--------------+------+------+-------------+--------+---------+------------+ | ... | batch_size | l1 | l2 | lr | iter | loss | accuracy | |-----+--------------+------+------+-------------+--------+---------+------------| | ... | 2 | 1 | 256 | 0.000668163 | 1 | 2.31479 | 0.0977 | | ... | 4 | 64 | 8 | 0.0331514 | 1 | 2.31605 | 0.0983 | | ... | 4 | 2 | 1 | 0.000150295 | 1 | 2.30755 | 0.1023 | | ... | 16 | 32 | 32 | 0.0128248 | 10 | 1.66912 | 0.4391 | | ... | 4 | 8 | 128 | 0.00464561 | 2 | 1.7316 | 0.3463 | | ... | 8 | 256 | 8 | 0.00031556 | 1 | 2.19409 | 0.1736 | | ... | 4 | 16 | 256 | 0.00574329 | 2 | 1.85679 | 0.3368 | | ... | 8 | 2 | 2 | 0.00325652 | 1 | 2.30272 | 0.0984 | | ... | 2 | 2 | 2 | 0.000342987 | 2 | 1.76044 | 0.292 | | ... | 4 | 64 | 32 | 0.003734 | 8 | 1.53101 | 0.4761 | +-----+--------------+------+------+-------------+--------+---------+------------+ Best trial config: {\u0027l1\u0027: 64, \u0027l2\u0027: 32, \u0027lr\u0027: 0.0037339984519545164, \u0027batch_size\u0027: 4} Best trial final validation loss: 1.5310075663924216 Best trial final validation accuracy: 0.4761 Best trial test set accuracy: 0.4737 \ub300\ubd80\ubd84\uc758 \uc2e4\ud5d8\uc740 \uc790\uc6d0 \ub0ad\ube44\ub97c \ub9c9\uae30 \uc704\ud574 \uc77c\ucc0d \uc911\ub2e8\ub418\uc5c8\uc2b5\ub2c8\ub2e4. \uac00\uc7a5 \uc88b\uc740 \uacb0\uacfc\ub97c \uc5bb\uc740 \uc2e4\ud5d8\uc740 47%\uc758 \uc815\ud655\ub3c4\ub97c \ub2ec\uc131\ud588\uc73c\uba70, \uc774\ub294 \ud14c\uc2a4\ud2b8\uc14b\uc5d0\uc11c \ud655\uc778\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\uac83\uc774 \uc804\ubd80\uc785\ub2c8\ub2e4! \uc774\uc81c \ud30c\uc774\ud1a0\uce58 \ubaa8\ub378\uc758 \ub9e4\uac1c\ubcc0\uc218\ub97c \uc870\uc815\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. Total running time of the script: (3 minutes 16.218 seconds) Download Jupyter notebook: hyperparameter_tuning_tutorial.ipynb Download Python source code: hyperparameter_tuning_tutorial.py Download zipped: hyperparameter_tuning_tutorial.zip",
"author": {
"@type": "Organization",
"name": "PyTorch Contributors",
"url": "https://pytorch.org"
},
"image": "../_static/img/pytorch_seo.png",
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "/beginner/hyperparameter_tuning_tutorial.html"
},
"datePublished": "2023-01-01T00:00:00Z",
"dateModified": "2023-01-01T00:00:00Z"
}
| article:modified_time | 2022-11-30T07:09:41+00:00 |
| og:type | article |
| og:site_name | PyTorch Tutorials KR |
| og:image | ../_static/img/pytorch_seo.png |
| og:image:alt | PyTorch Tutorials KR |
| og:ignore_canonical | true |
| docsearch:language | ko |
| docbuild:last-update | 2022년 11월 30일 |
| None | 2 |
| pytorch_project | tutorials |
Links:
Viewport: width=device-width, initial-scale=1