Update README.md
Browse files
README.md
CHANGED
@@ -49,6 +49,8 @@ tokenizer = open_clip.get_tokenizer('hf-hub:imageomics/bioclip-vit-b-16-inat-onl
|
|
49 |
|
50 |
Training was performed on 4 NVIDIA A100-80GB GPUs distributed over 1 node on [OSC's](https://www.osc.edu/) Ascend HPC Cluster with global batch size 16,384 for 2 days.
|
51 |
|
|
|
|
|
52 |
### Training Data
|
53 |
|
54 |
This model was trained on [iNat21](https://github.com/visipedia/inat_comp/tree/master/2021), which is a compilation of images matched to [Linnaean taxonomic rank](https://www.britannica.com/science/taxonomy/The-objectives-of-biological-classification) from kingdom through species. They are also matched with common (vernacular) name of the subject of the image where available.
|
|
|
49 |
|
50 |
Training was performed on 4 NVIDIA A100-80GB GPUs distributed over 1 node on [OSC's](https://www.osc.edu/) Ascend HPC Cluster with global batch size 16,384 for 2 days.
|
51 |
|
52 |
+
Based on Machine Learning Impact calculator presented in Lacoste et al. (2019), that's 33.16 kg of CO2 eq., or 134km driven by an average ICE car.
|
53 |
+
|
54 |
### Training Data
|
55 |
|
56 |
This model was trained on [iNat21](https://github.com/visipedia/inat_comp/tree/master/2021), which is a compilation of images matched to [Linnaean taxonomic rank](https://www.britannica.com/science/taxonomy/The-objectives-of-biological-classification) from kingdom through species. They are also matched with common (vernacular) name of the subject of the image where available.
|