riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-08-01 02:31:11 +06:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

change mention of decoder_input_ids to input_ids and same with decode_inputs_embeds (#26406)

Browse Source 
* change mention of decoder_input_ids to input_ids and same with decoder_input_embeds

* Style

---------

Co-authored-by: Lysandre <lysandre@huggingface.co>
This commit is contained in:
Tanishq Abraham 2023-09-28 01:15:48 -07:00 committed by GitHub
parent ba47efbfe4
commit 098c3f400c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/transformers/models/llama/modeling_llama.py
View File

@ -722,7 +722,7 @@ LLAMA_INPUTS_DOCSTRING = r"""
Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
[`PreTrainedTokenizer.__call__`] for details. [`PreTrainedTokenizer.__call__`] for details.
If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see If `past_key_values` is used, optionally only the last `input_ids` have to be input (see
`past_key_values`). `past_key_values`).
If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`] If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
@ -744,9 +744,9 @@ LLAMA_INPUTS_DOCSTRING = r"""
Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
blocks) that can be used (see `past_key_values` input) to speed up sequential decoding. blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't
don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids`
`decoder_input_ids` of shape `(batch_size, sequence_length)`. of shape `(batch_size, sequence_length)`.
inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*): inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
is useful if you want more control over how to convert `input_ids` indices into associated vectors than the is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
@ -843,13 +843,13 @@ class LlamaModel(LlamaPreTrainedModel):
# retrieve input_ids and inputs_embeds # retrieve input_ids and inputs_embeds
if input_ids is not None and inputs_embeds is not None: if input_ids is not None and inputs_embeds is not None:
raise ValueError("You cannot specify both decoder_input_ids and decoder_inputs_embeds at the same time") raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
elif input_ids is not None: elif input_ids is not None:
batch_size, seq_length = input_ids.shape batch_size, seq_length = input_ids.shape
elif inputs_embeds is not None: elif inputs_embeds is not None:
batch_size, seq_length, _ = inputs_embeds.shape batch_size, seq_length, _ = inputs_embeds.shape
else: else:
raise ValueError("You have to specify either decoder_input_ids or decoder_inputs_embeds") raise ValueError("You have to specify either input_ids or inputs_embeds")
seq_length_with_past = seq_length seq_length_with_past = seq_length
past_key_values_length = 0 past_key_values_length = 0