fixes

README.md

@@ -35,9 +35,9 @@ git clone https://github.com/ViTAE-Transformer/Remote-Sensing-RVSA.git
 ```
 3. **Append** the code for CVMMAE present in `utils_model/CVMMAE.py` to the file present in `Remote-Sensing-RVSA/MAEPretrain_SceneClassification/models_mae_vitae.py`
 
-4. Download pretrained satellite image encoder from - [Link](https://1drv.ms/u/s!AimBgYV7JjTlgUIde2jzcjrrWasP?e=gyLn29) and place inside folder `pretrained_models`.
+4. Download pretrained satellite image encoder from - [Link](https://1drv.ms/u/s!AimBgYV7JjTlgUIde2jzcjrrWasP?e=gyLn29) and place inside folder `pretrained_models`. You might get an error while loading this model. You need to set the option `kernel=3` in the file `Remote-Sensing-RVSA/MAEPretrain_SceneClassification/models_mae_vitae.py` in the class `MaskedAutoencoderViTAE`.
 
-5. Download all datasets and place inside folder `data`.
+5. Download all datasets, unzip them and place inside folder `data`.
 
 #### Installing Required Packages
 There are two options to setup your environment to be able to run all the functions in the repository:

config.py

@@ -24,6 +24,8 @@
 cfg.pretrain.overhead = edict()
 cfg.pretrain.overhead.img_size = (224, 224)
 cfg.pretrain.overhead.hori_flip_prob = 0.5
+cfg.pretrain.overhead.model = edict()
+cfg.pretrain.overhead.model.patch_size = 16
 
 
 cfg.pretrain.train = edict()

models.py

@@ -646,7 +646,7 @@ def __init__(self, train_dataset, val_dataset, **kwargs):
             img_size1=cfg.pretrain.ground.img_size,
             img_size2=cfg.pretrain.overhead.img_size,
             patch_size1=cfg.pretrain.ground.model.patch_size,
-            patch_size2=cfg.pretrain.overhead.patch_size,
+            patch_size2=cfg.pretrain.overhead.model.patch_size,
             in_chans=cfg.pretrain.ground.model.in_chans,
             embed_dim=cfg.pretrain.ground.model.embed_dim,
             depth=cfg.pretrain.ground.model.depth,