Fixes exceeding maximum sequence length when using generate().
Browse files- modeling_phi.py +16 -8
modeling_phi.py
CHANGED
@@ -481,7 +481,7 @@ def _update_kv_cache(kv: torch.FloatTensor, inference_params: InferenceParams, l
|
|
481 |
num_heads, head_dim = kv.shape[-2:]
|
482 |
|
483 |
if layer_idx not in inference_params.key_value_memory_dict:
|
484 |
-
|
485 |
inference_params.max_batch_size,
|
486 |
inference_params.max_seqlen,
|
487 |
2,
|
@@ -490,9 +490,6 @@ def _update_kv_cache(kv: torch.FloatTensor, inference_params: InferenceParams, l
|
|
490 |
dtype=kv.dtype,
|
491 |
device=kv.device,
|
492 |
)
|
493 |
-
inference_params.key_value_memory_dict[layer_idx] = kv_cache
|
494 |
-
else:
|
495 |
-
kv_cache = inference_params.key_value_memory_dict[layer_idx]
|
496 |
|
497 |
batch_start = inference_params.batch_size_offset
|
498 |
batch_end = batch_start + kv.shape[0]
|
@@ -500,9 +497,14 @@ def _update_kv_cache(kv: torch.FloatTensor, inference_params: InferenceParams, l
|
|
500 |
sequence_start = inference_params.seqlen_offset
|
501 |
sequence_end = sequence_start + kv.shape[1]
|
502 |
|
503 |
-
|
504 |
-
|
|
|
|
|
505 |
|
|
|
|
|
|
|
506 |
return kv
|
507 |
|
508 |
|
@@ -710,7 +712,6 @@ class MHA(nn.Module):
|
|
710 |
attention_mask: Optional[Union[torch.LongTensor, torch.BoolTensor]] = None,
|
711 |
**kwargs,
|
712 |
) -> Tuple[torch.FloatTensor, torch.FloatTensor]:
|
713 |
-
# TODO: Need an alternative way for dynamic control flow: torch.any(~attention_mask.bool())
|
714 |
if attention_mask is not None:
|
715 |
attention_mask = attention_mask.bool()
|
716 |
else:
|
@@ -863,6 +864,13 @@ class PhiPreTrainedModel(PreTrainedModel):
|
|
863 |
attention_mask: Optional[Union[torch.LongTensor, torch.BoolTensor]] = None,
|
864 |
**kwargs,
|
865 |
) -> Dict[str, Any]:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
866 |
if past_key_values is None or not (isinstance(past_key_values, InferenceParams)):
|
867 |
past_key_values = InferenceParams(
|
868 |
max_seqlen=self.config.n_positions,
|
@@ -874,7 +882,7 @@ class PhiPreTrainedModel(PreTrainedModel):
|
|
874 |
)
|
875 |
else:
|
876 |
# Assume that `past_key_values` has cached all tokens up to the last token in `input_ids`
|
877 |
-
past_key_values.seqlen_offset =
|
878 |
input_ids = input_ids[:, -1].unsqueeze(-1)
|
879 |
|
880 |
return {
|
|
|
481 |
num_heads, head_dim = kv.shape[-2:]
|
482 |
|
483 |
if layer_idx not in inference_params.key_value_memory_dict:
|
484 |
+
inference_params.key_value_memory_dict[layer_idx] = torch.empty(
|
485 |
inference_params.max_batch_size,
|
486 |
inference_params.max_seqlen,
|
487 |
2,
|
|
|
490 |
dtype=kv.dtype,
|
491 |
device=kv.device,
|
492 |
)
|
|
|
|
|
|
|
493 |
|
494 |
batch_start = inference_params.batch_size_offset
|
495 |
batch_end = batch_start + kv.shape[0]
|
|
|
497 |
sequence_start = inference_params.seqlen_offset
|
498 |
sequence_end = sequence_start + kv.shape[1]
|
499 |
|
500 |
+
# When the current sequence length is equal to or larger than the maximum sequence length,
|
501 |
+
# we need to roll the cache to the left and update it
|
502 |
+
if sequence_end >= inference_params.max_seqlen:
|
503 |
+
inference_params.key_value_memory_dict[layer_idx] = inference_params.key_value_memory_dict[layer_idx].roll(-(sequence_end - sequence_start), 1)
|
504 |
|
505 |
+
inference_params.key_value_memory_dict[layer_idx][batch_start:batch_end, sequence_start:sequence_end, ...] = kv
|
506 |
+
kv = inference_params.key_value_memory_dict[layer_idx][batch_start:batch_end, :sequence_end, ...]
|
507 |
+
|
508 |
return kv
|
509 |
|
510 |
|
|
|
712 |
attention_mask: Optional[Union[torch.LongTensor, torch.BoolTensor]] = None,
|
713 |
**kwargs,
|
714 |
) -> Tuple[torch.FloatTensor, torch.FloatTensor]:
|
|
|
715 |
if attention_mask is not None:
|
716 |
attention_mask = attention_mask.bool()
|
717 |
else:
|
|
|
864 |
attention_mask: Optional[Union[torch.LongTensor, torch.BoolTensor]] = None,
|
865 |
**kwargs,
|
866 |
) -> Dict[str, Any]:
|
867 |
+
# Truncate `input_ids` and `attention_mask` (if necessary) to prevent exceeding
|
868 |
+
# the maximum sequence length
|
869 |
+
if input_ids.shape[1] > self.config.n_positions:
|
870 |
+
input_ids = input_ids[:, -self.config.n_positions :]
|
871 |
+
if attention_mask is not None:
|
872 |
+
attention_mask = attention_mask[:, -self.config.n_positions :]
|
873 |
+
|
874 |
if past_key_values is None or not (isinstance(past_key_values, InferenceParams)):
|
875 |
past_key_values = InferenceParams(
|
876 |
max_seqlen=self.config.n_positions,
|
|
|
882 |
)
|
883 |
else:
|
884 |
# Assume that `past_key_values` has cached all tokens up to the last token in `input_ids`
|
885 |
+
past_key_values.seqlen_offset = input_ids.shape[1] - 1
|
886 |
input_ids = input_ids[:, -1].unsqueeze(-1)
|
887 |
|
888 |
return {
|