https://github.com/kempnerinstitute/intro-compute-march-2024
https://github.com/kempnerinstitute/intro-compute-march-2024
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/kempnerinstitute/intro-compute-march-2024
- Owner: KempnerInstitute
- Created: 2024-03-25T12:37:44.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2024-03-28T16:27:50.000Z (about 1 year ago)
- Last Synced: 2025-01-30T17:09:04.484Z (5 months ago)
- Language: Python
- Size: 35.2 KB
- Stars: 1
- Watchers: 6
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Kempner Institute Spring 2024 Compute Workshop Supplemental Material
Please see the [Workshop Page](TBD) for more information.
The supplementary material contains files for each run scenario. Below is a list of the folders along with a brief description of their purposes. In all models, we modified the first and the last layers of the ResNet-18 model to fit the CIFAR-10 dataset.
## SLURM Examples
- **SLURM Example 1**:
- Usefule SLURM commands
- Test Only Jobs
- Estimating job efficiency (CPU and Memory)
## GPU Examples
- **GPU Example 1**:
- Model: ResNet-18
- Dataset: CIFAR-10
- Resources: 1 GPU (A100)
- Job submission: Interactive
- **GPU Example 2**:
- Model: ResNet-18
- Dataset: CIFAR-10
- Resources: 1 GPU (A100)
- Job submission: Batch
- Integration: Weights & Biases (wandb.ai)
- **GPU Example 3**:
- Model: ResNet-18
- Dataset: CIFAR-10
- Resources: 1 GPU (A100)
- Job submission: Batch
- Integration: Weights & Biases (wandb.ai)
- Job Arrays
- **GPU Example 4**:
- Model: ResNet-50
- Dataset: CIFAR-10
- Resources: 2, 4 GPUs (A100)
- Job submission: Batch
- Integration: Weights & Biases (wandb.ai)
- **GPU Example 5**:
- Model: ResNet-50
- Dataset: Subset of imagenet
- Resources: 4 GPUs, 8 GPUs (two node) (A100)
- Job submission: Batch