more fixes to config

configs/train_decoder_config.test.json

@@ -9,7 +9,7 @@
                 "dim_mults": [1, 2, 4, 8],
                 "attn_dim_head": 16,
                 "attn_heads": 4,
-		"self_attn": [false, true, true, true]
+                "self_attn": [false, true, true, true]
             }
         ],
         "clip": {

dalle2_pytorch/__init__.py

@@ -1,10 +1,3 @@
-import torch
-from packaging import version
-
-if version.parse(torch.__version__) >= version.parse('2.0.0'):
-    from einops._torch_specific import allow_ops_in_compiled_graph
-    allow_ops_in_compiled_graph()
-
 from dalle2_pytorch.version import __version__
 from dalle2_pytorch.dalle2_pytorch import DALLE2, DiffusionPriorNetwork, DiffusionPrior, Unet, Decoder
 from dalle2_pytorch.dalle2_pytorch import OpenAIClipAdapter, OpenClipAdapter

dalle2_pytorch/train_configs.py

@@ -233,7 +233,7 @@ class UnetConfig(BaseModel):
     cond_on_text_encodings: Optional[bool] = None
     cond_dim: Optional[int] = None
     channels: int = 3
-    self_attn: ListOrTuple[int]
+    self_attn: ListOrTuple[bool]
     attn_dim_head: int = 32
     attn_heads: int = 16
     init_cross_embed: bool = True
@@ -245,7 +245,7 @@ class DecoderConfig(BaseModel):
     unets: ListOrTuple[UnetConfig]
     image_size: Optional[int] = None
     image_sizes: ListOrTuple[int] = None
-    clip: Optional[AdapterConfig]   # The clip model to use if embeddings are not provided
+    clip: Optional[AdapterConfig] = None   # The clip model to use if embeddings are not provided
     channels: int = 3
     timesteps: int = 1000
     sample_timesteps: Optional[SingularOrIterable[Optional[int]]] = None

dalle2_pytorch/version.py

@@ -1 +1 @@
-__version__ = '1.15.3'
+__version__ = '1.15.5'

setup.py

@@ -30,7 +30,7 @@ setup(
     'clip-anytorch>=2.5.2',
     'coca-pytorch>=0.0.5',
     'ema-pytorch>=0.0.7',
-    'einops>=0.6.1',
+    'einops>=0.7.0',
     'embedding-reader',
     'kornia>=0.5.4',
     'numpy',