Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
cjrd committed Feb 8, 2023
1 parent 2795a72 commit fcfded5
Showing 1 changed file with 28 additions and 3 deletions.
31 changes: 28 additions & 3 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,13 +1,36 @@
# Scale-MAE 🛰️

## Requirements and Setup ##
Set up your Python (`conda` suggested) environment in the same way as the [MAE](github.com/facebookresearch/mae) repository. Please note that the `timm==0.3.2` patch is still needed for Scale-MAE.
![image](https://user-images.githubusercontent.com/1455579/217665789-b46d6830-445f-4151-b7a4-a2152a81a8d1.png)


This repository provides a reimplementation of the code for [Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learning](https://arxiv.org/abs/2212.14532) (the original code was optimized for our distributed cluster).

```
@article{reed2022scale,
title={Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learning},
author={Reed, Colorado J and Gupta, Ritwik and Li, Shufan and Brockman, Sarah and Funk, Christopher and Clipp, Brian and Candido, Salvatore and Uyttendaele, Matt and Darrell, Trevor},
journal={arXiv preprint arXiv:2212.14532},
year={2022}
}
```

* This repo is a modification on the [MAE repo](https://github.com/facebookresearch/mae). Installation and preparation follow that repo ;-).

* As mentioned in the MAE repo, this repo is based on [`timm==0.3.2`](https://github.com/rwightman/pytorch-image-models), for which a [fix](https://github.com/rwightman/pytorch-image-models/issues/420#issuecomment-776459842) is needed to work with PyTorch 1.8.1+. In addition, install gdal, rasterio, and Shapely. This tends to work pretty well (but gdal is notoriously tricky):

```bash
conda install geopandas. # this should setup gdal correctly...
pip install rasterio shapely
```

In addition, install `GDAL`, `rasterio`, and `Shapely`.

## Pretraining ##
Datasets are defined by config files in `config`



# TOD FIX ALL OF THIS

```
python -m torch.distributed.launch --nproc_per_node=4 --master_port=11083 main_pretrain.py\
--batch_size 256 \
Expand Down Expand Up @@ -70,4 +93,6 @@ main_linprobe.py \
flag `--finetune` is enabled for fine-tuning and not enabled for linear probing


---

> Note: THIS SOFTWARE AND/OR DATA WAS DEPOSITED IN THE BAIR OPEN RESEARCH COMMONS REPOSITORY ON 2/8/23.

0 comments on commit fcfded5

Please sign in to comment.