Updated docs of forward in Idefics2ForConditionalGeneration with …

…correct `ignore_index` value (#30678) updated docs of `forward` in `Idefics2ForConditionalGeneration` with correct `ignore_index` value
huggingface · May 7, 2024 · e5f71ec · e5f71ec
1 parent 9c8979e
commit e5f71ec
Showing 1 changed file with 3 additions and 3 deletions.
diff --git a/src/transformers/models/idefics2/modeling_idefics2.py b/src/transformers/models/idefics2/modeling_idefics2.py
@@ -1776,9 +1776,9 @@ def forward(
         Args:
             labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
                 Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
-                config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
-                (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
-
+                config.vocab_size]` or `model.image_token_id` (where `model` is your instance of `Idefics2ForConditionalGeneration`).
+                Tokens with indices set to `model.image_token_id` are ignored (masked), the loss is only
+                computed for the tokens with labels in `[0, ..., config.vocab_size]`.
         Returns:
 
         Example: