|
--- |
|
license: apache-2.0 |
|
metrics: |
|
- accuracy |
|
pipeline_tag: image-classification |
|
tags: |
|
- fastai |
|
- astronomy |
|
--- |
|
|
|
## Astronomical Classifier |
|
|
|
Classify images of astronomical objects such as galaxies, nebulae, comets, asteroids, quasars, and star clusters. |
|
|
|
Built for the fast.ai [Practical Deep Learning Course](https://course.fast.ai/) by: |
|
- creating a custom dataset (less than 150 images per label) using [Bing search API](https://www.microsoft.com/en-us/bing/apis/bing-web-search-api); |
|
- augmenting the dataset; and, |
|
- fine tuning ResNet50 (1 + 3 epochs) in [paperspace.com](https://www.paperspace.com/). |
|
|
|
Try at: https://huggingface.co/spaces/dcarpintero/fastai-interstellar |
|
|
|
Two versions of the model are provided: |
|
|
|
### Class Model |
|
|
|
*This version [fastai-interstellar-class](https://huggingface.co/dcarpintero/fastai-interstellar-class/tree/main)* |
|
|
|
Classifies an image under an astronomy class: |
|
- `asteroid` |
|
- `comet` |
|
- `galaxy` |
|
- `nebulae` |
|
- `planet` |
|
- `quasar in space` |
|
- `star cluster` |
|
|
|
Accuracy: 84% |
|
|
|
---- |
|
|
|
### Object Model |
|
|
|
*Available at [fastai-interstellar-object](https://huggingface.co/dcarpintero/fastai-interstellar-object/tree/main)* |
|
|
|
Recognizes a (limited) set of **specific** astronomical objects: |
|
- `m31 andromeda` |
|
- `m33 triangulum` |
|
- `m81 bode` |
|
- `m82 cigar` |
|
- `ngc 1300` |
|
- `m104 sombrero` |
|
- `m51 whirlpool` |
|
- `m42 orion nebula` |
|
- `m17 omega nebula` |
|
- `m45 pleiades star cluster` |
|
|
|
Accuracy: 94.1% |
