cond drop prob for diffusion prior network should default to 0

README.md

@@ -1065,6 +1065,7 @@ Once built, images will be saved to the same directory the command is invoked
 - [ ] allow for unet to be able to condition non-cross attention style as well
 - [ ] for all model classes with hyperparameters that changes the network architecture, make it requirement that they must expose a config property, and write a simple function that asserts that it restores the object correctly
 - [ ] for both diffusion prior and decoder, all exponential moving averaged models needs to be saved and restored as well (as well as the step number)
+- [ ] read the paper, figure it out, and build it https://github.com/lucidrains/DALLE2-pytorch/issues/89
 
 ## Citations
 
@@ -1153,4 +1154,13 @@ Once built, images will be saved to the same directory the command is invoked
 }
 ```
 
+```bibtex
+@article{ho2021cascaded,
+    title   = {Cascaded Diffusion Models for High Fidelity Image Generation},
+    author  = {Ho, Jonathan and Saharia, Chitwan and Chan, William and Fleet, David J and Norouzi, Mohammad and Salimans, Tim},
+    journal = {arXiv preprint arXiv:2106.15282},
+    year    = {2021}
+}
+```
+
 *Creating noise from data is easy; creating data from noise is generative modeling.* - <a href="https://arxiv.org/abs/2011.13456">Yang Song's paper</a>

dalle2_pytorch/dalle2_pytorch.py

@@ -794,7 +794,7 @@ class DiffusionPriorNetwork(nn.Module):
         text_embed,
         text_encodings = None,
         mask = None,
-        cond_drop_prob = 0.2
+        cond_drop_prob = 0.
     ):
         batch, dim, device, dtype = *image_embed.shape, image_embed.device, image_embed.dtype
 

setup.py

@@ -10,7 +10,7 @@ setup(
       'dream = dalle2_pytorch.cli:dream'
     ],
   },
-  version = '0.2.35',
+  version = '0.2.36',
   license='MIT',
   description = 'DALL-E 2',
   author = 'Phil Wang',