Instructions to use iamcode6/dinov2-l-ccmt-mi300x with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- timm
How to use iamcode6/dinov2-l-ccmt-mi300x with timm:
import timm model = timm.create_model("hf-hub:iamcode6/dinov2-l-ccmt-mi300x", pretrained=True) - Notebooks
- Google Colab
- Kaggle
Download benchmark_results.json from iamcode6/dinov2-l-ccmt-mi300x: direct link, hf CLI and curl.
- Browser
- Download file 1.48 kB
-
https://huggingface.co/iamcode6/dinov2-l-ccmt-mi300x/resolve/main/benchmark_results.json
- Command line
-
hf download hf://iamcode6/dinov2-l-ccmt-mi300x/benchmark_results.json
-
curl -L -o benchmark_results.json https://huggingface.co/iamcode6/dinov2-l-ccmt-mi300x/resolve/main/benchmark_results.json
1.48 kB
| { | |
| "device": "", | |
| "rocm_hip": "7.0.51831-a3e329ad8", | |
| "torch": "2.9.0.dev20250821+rocm7.0.0.git125803b7", | |
| "model": "vit_large_patch14_dinov2.lvd142m", | |
| "img_size": 224, | |
| "amp_dtype": "torch.bfloat16", | |
| "results": [ | |
| { | |
| "model": "vit_large_patch14_dinov2.lvd142m", | |
| "img_size": 224, | |
| "batch_size": 256, | |
| "mode": "train", | |
| "steps": 50, | |
| "elapsed_s": 25.571, | |
| "images_per_sec": 500.6, | |
| "peak_vram_gb": 70.48 | |
| }, | |
| { | |
| "model": "vit_large_patch14_dinov2.lvd142m", | |
| "img_size": 224, | |
| "batch_size": 256, | |
| "mode": "eval", | |
| "steps": 50, | |
| "elapsed_s": 6.917, | |
| "images_per_sec": 1850.5, | |
| "peak_vram_gb": 133.44 | |
| }, | |
| { | |
| "model": "vit_large_patch14_dinov2.lvd142m", | |
| "img_size": 224, | |
| "batch_size": 512, | |
| "mode": "train", | |
| "steps": 50, | |
| "elapsed_s": 50.003, | |
| "images_per_sec": 512.0, | |
| "peak_vram_gb": 136.51 | |
| }, | |
| { | |
| "batch_size": 512, | |
| "mode": "eval", | |
| "error": "OOM" | |
| }, | |
| { | |
| "batch_size": 1024, | |
| "mode": "train", | |
| "error": "OOM" | |
| }, | |
| { | |
| "batch_size": 1024, | |
| "mode": "eval", | |
| "error": "OOM" | |
| } | |
| ], | |
| "baseline_p100_effnetb0": { | |
| "gpu": "Tesla P100-PCIE-16GB", | |
| "model": "efficientnet_b0", | |
| "img_size": 224, | |
| "batch_size": 32, | |
| "images_per_sec_train": null, | |
| "images_per_sec_eval": 353.0, | |
| "notes": "P100, eval pass only; training step time not cleanly logged" | |
| } | |
| } |