Update config

README.md

@@ -72,7 +72,7 @@ Applying to random noise produces
 # Associative Dictionary
 This is an attempt to store more than one image in a network based on text embedding and associated image. In principle it could also be a generative model if you ask for something not in the dictionary, but we'll see what happens. I'm using the [Pick-a-Pic](https://stability.ai/research/pick-a-pic) so you'll need to download those parquet files - the idea here is not (yet) to train on an enormous dataset but maybe use 10s to 100s of images.
 ```
-python3 examples/text_to_image.py batch_size=2 optimizer=sparse_lion
+python3 examples/text_to_image.py batch_size=2000 optimizer=sparse_lion max_epochs=10
 ```
 
 # Random Interpolation (A Generative Model)

config/generative_config.yaml

@@ -5,7 +5,7 @@ mlp:
   periodicity: null
   rescale_output: False
   scale: 2.0
-  width: 10
+  width: 100
   layers: 2
   segments: 2
 

high_order_implicit_representation/single_image_dataset.py

@@ -390,10 +390,6 @@ def gen_data(self):
         self.count+=1
 
     def __getitem__(self, idx):
-        # I'm totally ignoring the index
-        # ans = self.dataset()
-        # print('ans', ans)
-
         return next(self.generator)
 
 
@@ -423,9 +419,6 @@ def gen_data(self):
 
 
     def __getitem__(self, idx):
-        # I'm totally ignoring the index
-        # ans = self.dataset()
-        # print('ans', ans)
 
         return self.gen_data()
 
@@ -452,17 +445,6 @@ def train_dataset(self) -> Dataset:
     def test_dataset(self) -> Dataset:
         return self._test_dataset
 
-    def collate_fn_reset(self, dataset):
-        def collate(batch):
-            text = torch.vstack([row[0] for row in batch]) 
-            pos = torch.vstack([row[1] for row in batch])
-            rgb = torch.vstack([row[2] for row in batch])
-
-            #print('batch', batch)
-            return text, pos, rgb #torch.vstack(torch.from_numpy(batch[0])), torch.vstack(batch[1]), torch.vstack(batch[2])
-        dataset.reset()
-        return collate
-
     def train_dataloader(self) -> DataLoader:
         self._train_dataset.reset()
         return DataLoader(