Model dim (#53)

src/seamless_communication/models/unity/adaptor_block.py

@@ -22,7 +22,6 @@ from fairseq2.nn.transformer import (
     TransformerEncoderLayer,
     create_standard_layer_norm,
 )
-from fairseq2.nn.utils.module import check_model_dim
 from fairseq2.typing import DataType, Device
 from overrides import final as finaloverride
 from torch import Tensor
@@ -96,8 +95,6 @@ class UnitYEncoderAdaptor(TransformerEncoder):
 
         self.layer_norm = layer_norm_factory(model_dim, device=device, dtype=dtype)
 
-        check_model_dim(self)
-
     @finaloverride
     def forward(
         self,
@@ -241,8 +238,6 @@ class UnitYTransformerAdaptorLayer(TransformerEncoderLayer):
         else:
             self.register_module("ffn_dropout", None)
 
-        check_model_dim(self)
-
     @finaloverride
     def forward(
         self,

src/seamless_communication/models/unity/model.py

@@ -14,7 +14,6 @@ from fairseq2.nn.incremental_state import IncrementalStateBag
 from fairseq2.nn.padding import PaddingMask
 from fairseq2.nn.projection import Projection
 from fairseq2.nn.transformer import TransformerDecoder, TransformerEncoder
-from fairseq2.nn.utils.module import check_model_dim
 from overrides import final as finaloverride
 from torch import Tensor
 from torch.nn import Module
@@ -94,8 +93,6 @@ class UnitYModel(EncoderDecoderModel):
 
         self.pad_idx = pad_idx
 
-        check_model_dim(self)
-
     @finaloverride
     def encode(
         self, seqs: Tensor, padding_mask: Optional[PaddingMask]
@@ -189,7 +186,6 @@ class UnitYX2TModel(EncoderDecoderModel):
         self.decoder = decoder
         self.final_proj = final_proj
         self.pad_idx = pad_idx
-        check_model_dim(self)
 
     @finaloverride
     def encode(

src/seamless_communication/models/unity/nar_decoder_layer.py

@@ -18,7 +18,6 @@ from fairseq2.nn.transformer import (
 from fairseq2.nn.incremental_state import IncrementalStateBag
 from fairseq2.nn.padding import PaddingMask, apply_padding_mask
 from fairseq2.nn.transformer import create_standard_layer_norm
-from fairseq2.nn.utils.module import check_model_dim
 from fairseq2.typing import DataType, Device, finaloverride
 
 
@@ -166,8 +165,6 @@ class NARTransformerDecoderLayer(TransformerDecoderLayer):
             model_dim, device=device, dtype=dtype
         )
 
-        check_model_dim(self)
-
     @finaloverride
     def forward(
         self,

src/seamless_communication/models/wav2vec2_chunk/encoder.py

@@ -9,7 +9,6 @@ from typing import Iterable, Optional, Tuple, final
 from torch import Tensor
 from torch.nn import Dropout
 
-from fairseq2.nn.utils.module import check_model_dim
 from fairseq2.nn.module_list import ModuleList
 from fairseq2.nn.normalization import LayerNorm
 from fairseq2.nn.padding import PaddingMask
@@ -80,8 +79,6 @@ class ChunkTransformerEncoder(TransformerEncoder):
 
         self.layers = layer_list
 
-        check_model_dim(self)
-
     @finaloverride
     def forward(
         self,