Upload folder using huggingface_hub

README.md

@@ -23,7 +23,7 @@ model-index:
 
 # EgaraNet — Illustration Style Embedding Model
 
-EgaraNet is an embedding model that encodes the artistic style of illustrations into 1024-dimensional L2-normalized vectors. It was trained on approximately **1.2 million illustrations** from around **12,000 artists**, learning to produce embeddings where illustrations by the same artist are close together in the vector space.
+EgaraNet is an embedding model that encodes the artistic style of illustrations into 1024-dimensional L2-normalized vectors. It was trained on approximately **1.2 million illustrations** from around **12,000 artists**, learning to produce embeddings where illustrations by the same artist are close together in the vector space. Produced under Article 30-4 of the Japanese Copyright Act.
 
 ## Model Description
 
@@ -34,7 +34,6 @@ EgaraNet is an embedding model that encodes the artistic style of illustrations
 | **Input** | RGB images, any resolution (multiples of 16 recommended) |
 | **Output** | L2-normalized style embedding vector |
 | **Training Data** | ~1.2M illustrations from ~12K artists |
-| **Legal Basis** | Article 30-4 of the Japanese Copyright Act |
 
 ### Architecture
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9b0d29d8f3bca26aa2429d3a14684e90b4298aafb8f039dc078d16518f8c073
+oid sha256:bc07899ef37fc10821834d26888d52e1d6e5cc335f2e9aaf4a71c94e5f3c8ddb
 size 1389584856

modeling_egara_net.py

@@ -165,7 +165,7 @@ class StyleNet(nn.Module):
             else nn.Linear(input_dim, hidden_dim)
         )
 
-        self.tab_layers = nn.ModuleList([
+        self.ttb_layers = nn.ModuleList([
             TransposedTransformerBlock(
                 dim=hidden_dim,
                 num_heads=config.ttb_num_heads,
@@ -187,7 +187,7 @@ class StyleNet(nn.Module):
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
         x = self.input_proj(x)
-        for layer in self.tab_layers:
+        for layer in self.ttb_layers:
             x = layer(x)
         x = self.attn_pool(x)
         x = self.head(x)