Remove inplace

torchscale/architecture/decoder.py

@@ -31,7 +31,7 @@ class DecoderLayer(nn.Module):
         super().__init__()
         self.args = args
         self.embed_dim = args.decoder_embed_dim
-        self.dropout_module = torch.nn.Dropout(args.dropout, inplace=True)
+        self.dropout_module = torch.nn.Dropout(args.dropout)
 
         if args.drop_path_rate > 0:
             drop_path_prob = np.linspace(0, args.drop_path_rate, args.decoder_layers)[
@@ -217,7 +217,7 @@ class Decoder(nn.Module):
         super().__init__(**kwargs)
         self.args = args
 
-        self.dropout_module = torch.nn.Dropout(args.dropout, inplace=True)
+        self.dropout_module = torch.nn.Dropout(args.dropout)
 
         embed_dim = args.decoder_embed_dim
         self.embed_dim = embed_dim

torchscale/architecture/encoder.py

@@ -29,7 +29,7 @@ class EncoderLayer(nn.Module):
         self.embed_dim = args.encoder_embed_dim
         self.self_attn = self.build_self_attention(self.embed_dim, args)
         self.self_attn_layer_norm = MultiwayWrapper(args, LayerNorm(self.embed_dim, eps=args.layernorm_eps))
-        self.dropout_module = torch.nn.Dropout(args.dropout, inplace=True)
+        self.dropout_module = torch.nn.Dropout(args.dropout)
 
         if args.drop_path_rate > 0:
             drop_path_prob = np.linspace(0, args.drop_path_rate, args.encoder_layers)[
@@ -174,7 +174,7 @@ class Encoder(nn.Module):
         self.args = args
         super().__init__(**kwargs)
 
-        self.dropout_module = torch.nn.Dropout(args.dropout, inplace=True)
+        self.dropout_module = torch.nn.Dropout(args.dropout)
 
         embed_dim = args.encoder_embed_dim
         self.embed_scale = 1.0 if args.no_scale_embedding else math.sqrt(embed_dim)

torchscale/component/feedforward_network.py

@@ -109,10 +109,8 @@ class FeedForwardNetwork(nn.Module):
         super().__init__()
         self.embed_dim = embed_dim
         self.activation_fn = get_activation_fn(activation=str(activation_fn))
-        self.activation_dropout_module = torch.nn.Dropout(
-            activation_dropout, inplace=True
-        )
-        self.dropout_module = torch.nn.Dropout(dropout, inplace=True)
+        self.activation_dropout_module = torch.nn.Dropout(activation_dropout)
+        self.dropout_module = torch.nn.Dropout(dropout)
         self.fc1 = nn.Linear(self.embed_dim, ffn_dim)
         self.fc2 = nn.Linear(ffn_dim, self.embed_dim)
         self.ffn_layernorm = LayerNorm(ffn_dim, eps=layernorm_eps) if subln else None

torchscale/component/multihead_attention.py

@@ -48,7 +48,7 @@ class MultiheadAttention(nn.Module):
             if subln and self.self_attention
             else None
         )
-        self.dropout_module = torch.nn.Dropout(dropout, inplace=True)
+        self.dropout_module = torch.nn.Dropout(dropout)
         self.xpos = (
             XPOS(self.head_dim, args.xpos_scale_base)
             if args.xpos_rel_pos and self.self_attention