---
library_name: transformers
tags:
- image-geolocation
- geolocation
- geography
- geoguessr
- multi-modal
license: cc-by-nc-4.0
language:
- en
base_model: openai/clip-vit-large-patch14-336
pipeline_tag: zero-shot-image-classification
---

# Model Card for Thesis-CLIP-geoloc-continent

CLIP-ViT model fine-tuned for image geolocation. Optimized for queries at continent-level.


## Model Details

### Model Description


- **Developed by:** [jrheiner](https://huggingface.co/jrheiner)
<!-- - **Funded by [optional]:** [More Information Needed] -->
<!-- - **Shared by [optional]:** [More Information Needed] -->
- **Model type:** CLIP-ViT
- **Language(s) (NLP):** English
- **License:** Creative Commons Attribution Non Commercial 4.0
- **Finetuned from model: [openai/clip-vit-large-patch14-336](https://huggingface.co/openai/clip-vit-large-patch14-336)** 

### Model Sources

<!-- Provide the basic links for the model. -->

- **Repository:** https://github.com/jrheiner/thesis-appendix
<!-- - **Paper:** [More Information Needed] -->
- **Demo:** [Image Geolocation Demo Space](https://huggingface.co/spaces/jrheiner/thesis-demo)


## How to Get Started with the Model

```python
from PIL import Image
import requests
from transformers import CLIPProcessor, CLIPModel

model = CLIPModel.from_pretrained("jrheiner/thesis-clip-geoloc-continent")
processor = CLIPProcessor.from_pretrained("jrheiner/thesis-clip-geoloc-continent")

url = "https://huggingface.co/spaces/jrheiner/thesis-demo/resolve/main/kerger-test-images/Oceania_Australia_-32.947127313081_151.47903359833_kerger.jpg"
image = Image.open(requests.get(url, stream=True).raw)
choices = ["North America", "Africa", "Asia", "Oceania", "South America", "Europe"]
inputs = processor(text=choices, images=image, return_tensors="pt", padding=True)
outputs = model(**inputs)
logits_per_image = outputs.logits_per_image # this is the image-text similarity score
probs = logits_per_image.softmax(dim=1) # we can take the softmax to get the label probabilities
```

## Training Details

The model was fine-tuned on 177 270 images (29 545 per continent) sourced from Mapillary.