Update modeling_reasonir_8b.py

modeling_reasonir_8b.py

@@ -26,22 +26,22 @@ import torch.nn.functional as F
 import torch.utils.checkpoint
 from torch import nn
 
-from ...activations import ACT2FN
-from ...cache_utils import Cache, DynamicCache, StaticCache
-from ...modeling_attn_mask_utils import _prepare_4d_causal_attention_mask, _prepare_4d_causal_attention_mask_for_sdpa, _prepare_4d_attention_mask, _prepare_4d_attention_mask_for_sdpa
-from ...generation import GenerationMixin
-from ...modeling_attn_mask_utils import AttentionMaskConverter
-from ...modeling_outputs import (
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache, StaticCache
+from transformers.modeling_attn_mask_utils import _prepare_4d_causal_attention_mask, _prepare_4d_causal_attention_mask_for_sdpa, _prepare_4d_attention_mask, _prepare_4d_attention_mask_for_sdpa
+from transformers.generation import GenerationMixin
+from transformers.modeling_attn_mask_utils import AttentionMaskConverter
+from transformers.modeling_outputs import (
     BaseModelOutputWithPast,
     CausalLMOutputWithPast,
     QuestionAnsweringModelOutput,
     SequenceClassifierOutputWithPast,
     TokenClassifierOutput,
 )
-from ...modeling_rope_utils import ROPE_INIT_FUNCTIONS
-from ...modeling_utils import PreTrainedModel
-from ...pytorch_utils import ALL_LAYERNORM_LAYERS
-from ...utils import (
+from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS
+from transformers.modeling_utils import PreTrainedModel
+from transformers.pytorch_utils import ALL_LAYERNORM_LAYERS
+from transformers.utils import (
     add_code_sample_docstrings,
     add_start_docstrings,
     add_start_docstrings_to_model_forward,
@@ -201,8 +201,8 @@ class LlamaDynamicNTKScalingRotaryEmbedding(LlamaRotaryEmbedding):
 
 def rotate_half(x):
     """Rotates half the hidden dims of the input."""
-    x1 = x[..., : x.shape[-1] // 2]
-    x2 = x[..., x.shape[-1] // 2 :]
+    x1 = x[transformers., : x.shape[-1] // 2]
+    x2 = x[transformers., x.shape[-1] // 2 :]
     return torch.cat((-x2, x1), dim=-1)
 
 
@@ -1313,9 +1313,9 @@ class LlamaForCausalLM(LlamaPreTrainedModel, GenerationMixin):
         r"""
         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, ...,
+                Labels for computing the masked language modeling loss. Indices should either be in `[0, transformers.,
                 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]`.
+                (masked), the loss is only computed for the tokens with labels in `[0, transformers., config.vocab_size]`.
 
             num_logits_to_keep (`int`, *optional*):
                 Calculate logits for the last `num_logits_to_keep` tokens. If `0`, calculate logits for all
@@ -1433,7 +1433,7 @@ class LlamaForSequenceClassification(LlamaPreTrainedModel):
     ) -> Union[Tuple, SequenceClassifierOutputWithPast]:
         r"""
         labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, transformers.,
             config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
             `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
         """
@@ -1628,7 +1628,7 @@ class LlamaForTokenClassification(LlamaPreTrainedModel):
     ) -> Union[Tuple, TokenClassifierOutput]:
         r"""
         labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
-            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, transformers.,
             config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
             `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
         """
@@ -1662,4 +1662,4 @@ class LlamaForTokenClassification(LlamaPreTrainedModel):
             logits=logits,
             hidden_states=outputs.hidden_states,
             attentions=outputs.attentions,
-        )
+        )