TRF011

forward() must not read non-nn.Module attributes off submodules: pipeline parallelism may replace them with Identity.

Default	Enabled
Scope	All models
Source	mlinter/trf011.py
Show in terminal	mlinter --rule TRF011

What it does

In forward() of PreTrainedModel subclasses, flags submodule attribute accesses torch.nn.Identity would not have: on loop variables over self.layers, and self.<submodule>.<attr> where <attr> is not a standard nn.Module attribute.

Why is this bad?

Pipeline parallelism may replace any submodule with torch.nn.Identity, so reading a custom attribute (e.g. decoder_layer.attention_type) off it raises AttributeError at runtime. Read per-layer metadata from self.config.

Example

 def forward(self, ...):
-    for decoder_layer in self.layers:
+    for i, decoder_layer in enumerate(self.layers):
         hidden_states = decoder_layer(
             hidden_states,
-            attention_mask=causal_mask_mapping[decoder_layer.attention_type],
+            attention_mask=causal_mask_mapping[self.config.layer_types[i]],
         )

Suppressing this rule

Add a # trf-ignore: TRF011 comment on the flagged line or the line directly above it. See Suppressing rules for whole-file directives and when a suppression is the wrong answer.