Resnet50 Cifar100 A100
A demo model from the Data & Impact Accounting (DIA) lab. It performs image classification (CIFAR-100) via full fine-tune, with the base model microsoft/resnet-50, trained on NVIDIA A100.
Source du modèle
- Type
- Other
- Taille du fichier
- 90,8 MB
- Modèle de base
- microsoft/resnet-50
- Version
- vRevision 0930d07
- Mis à jour
- 17 sept. 2026
- Téléchargements
- 0
- Source
- Hugging Face
Description de la source
ResNet-50 — CIFAR-100 (NVIDIA A100)
A demo model from the Data & Impact Accounting (DIA) lab. It performs
image classification (CIFAR-100) via full fine-tune, with the base model microsoft/resnet-50, trained on
NVIDIA A100.
The point of this repo is not the model itself but its dia_report — a
standardized record of the energy, carbon, and water used to train it, embedded
in this card's metadata.
This footprint feeds the DIA dashboard, which rolls up a base model and all its derivatives to show the cumulative carbon, water, and energy cost of a model family.
Training footprint
| Metric | Value |
|---|---|
| Hardware | 1× NVIDIA A100-SXM4-80GB |
| Compute | 0.6684 GPU-hours |
| Energy | 0.1369 (measured) kWh |
| Carbon | 0.0089 (measured) kgCO2eq |
| Water | 0.246–0.548 (estimated-from-default-wue) L |
| Grid region | ca-on |
Energy and carbon are measured with CodeCarbon; water is estimated from a default water-usage-effectiveness range. Carbon uses the local grid's intensity (Ontario, ~0.03 kgCO2eq/kWh).
Reproduce
REPO=DIA-MVP/resnet50-cifar100-a100 python scripts/train_resnet50_cifar.py
Links
- Footprint table (dataset): DIA-MVP/dia-state-lab-2026
- Project / paper: ai-impact-accounting
- Lab workflow: see
LAB.mdin the repo
Sources
1 source
Sources
1 sourceVérifié 2 sept.
Artefacts du modèle
1 artefact
Artefacts du modèle
1 artefactExtraits de sources
2 extraits
Extraits de sources
2 extraits--- dia_version: '0.1' dia_report: scope: incremental lineage: - model: microsoft/resnet-50 relation: finetune compute: hardware: gpu: NVIDIA A100-SXM4-80GB count: 1 duration_gpu_hours: 0.6684 footprint: energy_kwh: value: 0.1369 quality: measured carbon_kgco2eq: value: 0.0089 quality: measured water_liters: value: - 0.246 - 0.548 quality: estimated-from-default-wue context: region: ca-on carbon_intensity: 0.03 wue_l_per_kwh: - 1.8 - 4.0 tool: codecarbon tags: - dia - carbon-footprint - energy-efficiency - sustainability license: apache-2.0 pipeline_tag: image-classification base_model: microsoft/resnet-50 --- # ResNet-50 — CIFAR-100 (NVIDIA A100) A demo model from the **Data & Impact Accounting (DIA)** lab. It performs image classification (CIFAR-100) via **full fine-tune**, with the base model `microsoft/resnet-50`, trained on **NVIDIA A100**. The point of this repo is not the model itself but its **`dia_report`** — a standardized record of the energy, carbon, and water used to train it, embedded in this card's metadata. This footprint feeds the DIA dashboard, which rolls up a base model and all its derivatives to show the **cumulative** carbon, water, and energy cost of a model family. ## Training footprint | Metric | Value | |---|---| | Hardware | 1× NVIDIA A100-SXM4-80GB | | Compute | 0.6684 GPU-hours | | Energy | 0.1369 (measured) kWh | | Carbon | 0.0089 (measured) kgCO₂eq | | Water | 0.246–0.548 (estimated-from-default-wue) L | | Grid region | ca-on | *Energy and carbon are measured with [CodeCarbon](https://github.com/mlco2/codecarbon); water is estimated from a default water-usage-effectiveness range. Carbon uses the local grid's intensity (Ontario, ~0.03 kgCO₂eq/kWh).* ## Reproduce ```bash REPO=DIA-MVP/resnet50-cifar100-a100 python scripts/train_resnet50_cifar.py ``` ## Links - **Footprint table (dataset):** [DIA-MVP/dia-state-lab-2026](https://huggingface.co/datasets/DIA-MVP/dia-state-lab-2026) - **Project / paper:** [ai-impact-accounting](https://github.com/VectorInstitute/ai-impact-accounting) - **Lab workflow:** see `LAB.md` in the repo
Source context: 0 downloads · 0 likes · Pipeline image-classification · Repo DIA-MVP/resnet50-cifar100-a100
