revert restriction for classifier free guidance for diffusion prior, given @crowsonkb advice

Added metric to track image_embed vs predicted_image_embed

README.md

@@ -981,6 +981,8 @@ Once built, images will be saved to the same directory the command is invoked
 - [ ] make sure FILIP works with DALL-E2 from x-clip https://arxiv.org/abs/2111.07783
 - [ ] make sure resnet hyperparameters can be configurable across unet depth (groups and expansion factor)
 - [ ] offer save / load methods on the trainer classes to automatically take care of state dicts for scalers / optimizers / saving versions and checking for breaking changes
+- [ ] offer setting in diffusion prior to split time and image embeddings into multiple tokens, configurable, for more surface area during attention
+- [ ] bring in skip-layer excitatons (from lightweight gan paper) to see if it helps for either decoder of unet or vqgan-vae training
 
 ## Citations
 

dalle2_pytorch/dalle2_pytorch.py

@@ -800,7 +800,7 @@ class DiffusionPrior(BaseGaussianDiffusion):
         image_size = None,
         image_channels = 3,
         timesteps = 1000,
-        cond_drop_prob = 0.2,
+        cond_drop_prob = 0.,
         loss_type = "l1",
         predict_x_start = True,
         beta_schedule = "cosine",

setup.py

@@ -10,7 +10,7 @@ setup(
       'dream = dalle2_pytorch.cli:dream'
     ],
   },
-  version = '0.1.8',
+  version = '0.1.10',
   license='MIT',
   description = 'DALL-E 2',
   author = 'Phil Wang',

train_diffusion_prior.py

@@ -93,6 +93,8 @@ def report_cosine_sims(diffusion_prior, image_reader, text_reader, train_set_siz
             text_embed, predicted_image_embeddings).cpu().numpy()
         unrelated_similarity = cos(
             text_embed, predicted_unrelated_embeddings).cpu().numpy()
+        predicted_img_similarity = cos(
+            test_image_embeddings, predicted_image_embeddings).cpu().numpy()
 
         wandb.log(
             {"CosineSimilarity(text_embed,image_embed)": np.mean(original_similarity)})
@@ -100,6 +102,8 @@ def report_cosine_sims(diffusion_prior, image_reader, text_reader, train_set_siz
             predicted_similarity)})
         wandb.log({"CosineSimilarity(text_embed,predicted_unrelated_embed)": np.mean(
             unrelated_similarity)})
+        wandb.log({"CosineSimilarity(image_embed,predicted_image_embed)": np.mean(
+            predicted_img_similarity)})
 
     return np.mean(predicted_similarity - original_similarity)