broccoli-ml 11.0.0__tar.gz → 12.0.0__tar.gz

{broccoli_ml-11.0.0 → broccoli_ml-12.0.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: broccoli-ml
-Version: 11.0.0
+Version: 12.0.0
 Summary: Some useful Pytorch models, circa 2025
 License: MIT
 Author: Nicholas Bailey

{broccoli_ml-11.0.0 → broccoli_ml-12.0.0}/broccoli/transformer.py

@@ -21,29 +21,6 @@ except ImportError:
     FLASH_ATTN = False
 
 
-class LayerScale(nn.Module):
-    def __init__(self, dim, decay=False, init_values=1e-4):
-        super().__init__()
-        self.dim = dim
-        self.decay = decay
-        self.init_values = init_values
-        self.reset_parameters()
-
-    def forward(self, x):
-        if self.decay:
-            return x * self.scale
-        else:
-            return x * self.nondecay_scale
-
-    def reset_parameters(self):
-        if self.decay:
-            self.scale = nn.Parameter(self.init_values * torch.ones(self.dim))
-            self.nondecay_scale = None
-        else:
-            self.nondecay_scale = nn.Parameter(self.init_values * torch.ones(self.dim))
-            self.scale = None
-
-
 def drop_path(
     x, drop_prob: float = 0.0, training: bool = False, scale_by_keep: bool = True
 ):
@@ -527,7 +504,6 @@ class TransformerBlock(nn.Module):
         post_norm=False,
         normformer=False,
         checkpoint_ff=True,
-        layerscale=True,
     ):
         """
         Args:
@@ -556,14 +532,6 @@ class TransformerBlock(nn.Module):
             self.post_attention_norm = nn.LayerNorm(d_model)
             self.post_mlp_norm = nn.LayerNorm(d_model)
 
-        self.layerscale = layerscale
-        if layerscale:
-            self.layerscale1 = LayerScale(d_model)
-            self.layerscale2 = LayerScale(d_model)
-        else:
-            self.layerscale1 = nn.Identity()
-            self.layerscale2 = nn.Identity()
-
         if relative_position_embedding:
             max_freq = int(max(source_size) / 2)  # Suggested by Gemini!
             if d_model < 16:
@@ -630,9 +598,7 @@ class TransformerBlock(nn.Module):
         else:
             process_x = x
 
-        processed = self.drop_path(
-            self.layerscale1(self.attn(process_x, process_x, process_x))
-        )
+        processed = self.drop_path(self.attn(process_x, process_x, process_x))
 
         if self.normformer:
             processed = self.normformer_norm(processed)
@@ -647,7 +613,7 @@ class TransformerBlock(nn.Module):
         else:
             process_x = x
 
-        x = x + self.drop_path(self.layerscale2(self.ff(process_x)))
+        x = x + self.drop_path(self.ff(process_x))
 
         if self.post_norm:
             x = self.post_mlp_norm(x)
@@ -682,10 +648,6 @@ class TransformerBlock(nn.Module):
         self.attn.reset_parameters()
         self.ff.reset_parameters()
 
-        if self.layerscale:
-            self.layerscale1.reset_parameters()
-            self.layerscale2.reset_parameters()
-
 
 class TransformerEncoder(nn.Module):
     """
@@ -722,7 +684,6 @@ class TransformerEncoder(nn.Module):
         normformer=False,
         msa_scaling="d",
         checkpoint_ff=True,
-        layerscale=True,
     ):
         """
         Args:
@@ -757,12 +718,6 @@ class TransformerEncoder(nn.Module):
         self._utility_tokens = utility_tokens
         self.return_utility_tokens = return_utility_tokens
 
-        if layerscale:
-            rope_and_ape = absolute_position_embedding and relative_position_embedding
-            self.position_layerscale = LayerScale(d_model, decay=rope_and_ape)
-        else:
-            self.position_layerscale = None
-
         # Initialise utility tokens with normal init, like usual Pytorch embeddings
         if self._utility_tokens:
             self._utility_token_embedding = nn.Parameter(
@@ -827,7 +782,6 @@ class TransformerEncoder(nn.Module):
                     post_norm=post_norm,
                     normformer=normformer,
                     checkpoint_ff=checkpoint_ff,
-                    layerscale=layerscale,
                 )
                 for i in range(n_layers)
             ]
@@ -855,8 +809,6 @@ class TransformerEncoder(nn.Module):
                     0
                 )  # to shape (1, seq_len) to broadcast over batch
             )
-            if self.position_layerscale is not None:
-                position_embedding = self.position_layerscale(position_embedding)
             x += position_embedding
 
         return x

{broccoli_ml-11.0.0 → broccoli_ml-12.0.0}/broccoli/vit.py

@@ -187,7 +187,6 @@ class ViTEncoder(nn.Module):
         transformer_msa_dropout=0.1,
         transformer_stochastic_depth=0.1,
         transformer_checkpoint_ff=True,
-        transformer_layerscale=True,
         linear_module=nn.Linear,
     ):
         super().__init__()
@@ -353,7 +352,6 @@ class ViTEncoder(nn.Module):
                 normformer=transformer_normformer,
                 post_norm=transformer_post_norm,
                 checkpoint_ff=transformer_checkpoint_ff,
-                layerscale=transformer_layerscale,
             )
         else:
             self.transformer = nn.Identity()
@@ -489,7 +487,6 @@ class ViT(nn.Module):
         transformer_msa_dropout=0.1,
         transformer_stochastic_depth=0.1,
         transformer_checkpoint_ff=True,
-        transformer_layerscale=True,
         head=SequencePoolClassificationHead,
         batch_norm_logits=True,
         logit_projection_layer=nn.Linear,
@@ -562,7 +559,6 @@ class ViT(nn.Module):
             transformer_msa_dropout=transformer_msa_dropout,
             transformer_stochastic_depth=transformer_stochastic_depth,
             transformer_checkpoint_ff=transformer_checkpoint_ff,
-            transformer_layerscale=transformer_layerscale,
             linear_module=linear_module,
         )
 

{broccoli_ml-11.0.0 → broccoli_ml-12.0.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "broccoli-ml"
-version = "11.0.0"
+version = "12.0.0"
 description = "Some useful Pytorch models, circa 2025"
 authors = [
     {name = "Nicholas Bailey"}