diff --git a/src/transformers/models/align/modeling_align.py b/src/transformers/models/align/modeling_align.py index 063b3312ccc5..c226a3b36ac6 100644 --- a/src/transformers/models/align/modeling_align.py +++ b/src/transformers/models/align/modeling_align.py @@ -19,7 +19,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/altclip/modeling_altclip.py b/src/transformers/models/altclip/modeling_altclip.py index 8b98548d0bfe..61468141c570 100755 --- a/src/transformers/models/altclip/modeling_altclip.py +++ b/src/transformers/models/altclip/modeling_altclip.py @@ -20,7 +20,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from ...activations import ACT2FN from ...modeling_layers import GradientCheckpointingLayer diff --git a/src/transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py b/src/transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py index 516dc4187885..c445fbb0e36d 100644 --- a/src/transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py +++ b/src/transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py @@ -17,7 +17,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/autoformer/modeling_autoformer.py b/src/transformers/models/autoformer/modeling_autoformer.py index efa952a5a28b..fe11fc4c4860 100644 --- a/src/transformers/models/autoformer/modeling_autoformer.py +++ b/src/transformers/models/autoformer/modeling_autoformer.py @@ -22,7 +22,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/bamba/modular_bamba.py b/src/transformers/models/bamba/modular_bamba.py index 52814930a172..f2495b446aa5 100644 --- a/src/transformers/models/bamba/modular_bamba.py +++ b/src/transformers/models/bamba/modular_bamba.py @@ -22,7 +22,6 @@ from typing import Optional, TypedDict, Union import torch -import torch.utils.checkpoint from torch import nn from transformers.activations import ACT2FN diff --git a/src/transformers/models/bart/modeling_bart.py b/src/transformers/models/bart/modeling_bart.py index 55c7654fe2e5..0a1f2451cff1 100755 --- a/src/transformers/models/bart/modeling_bart.py +++ b/src/transformers/models/bart/modeling_bart.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/beit/modeling_beit.py b/src/transformers/models/beit/modeling_beit.py index 09c887bcd2b4..cb4e0d712651 100755 --- a/src/transformers/models/beit/modeling_beit.py +++ b/src/transformers/models/beit/modeling_beit.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/bert/modeling_bert.py b/src/transformers/models/bert/modeling_bert.py index 20edbf6383c5..b9238d8bb071 100755 --- a/src/transformers/models/bert/modeling_bert.py +++ b/src/transformers/models/bert/modeling_bert.py @@ -22,7 +22,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/bert_generation/modeling_bert_generation.py b/src/transformers/models/bert_generation/modeling_bert_generation.py index f29d22d06f83..4be87a0cd544 100755 --- a/src/transformers/models/bert_generation/modeling_bert_generation.py +++ b/src/transformers/models/bert_generation/modeling_bert_generation.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/big_bird/modeling_big_bird.py b/src/transformers/models/big_bird/modeling_big_bird.py index 20a5a08c246a..f42b1eeaeeb1 100755 --- a/src/transformers/models/big_bird/modeling_big_bird.py +++ b/src/transformers/models/big_bird/modeling_big_bird.py @@ -21,7 +21,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/biogpt/modular_biogpt.py b/src/transformers/models/biogpt/modular_biogpt.py index 5753e066913f..001c1de65756 100644 --- a/src/transformers/models/biogpt/modular_biogpt.py +++ b/src/transformers/models/biogpt/modular_biogpt.py @@ -19,7 +19,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss from ...activations import ACT2FN diff --git a/src/transformers/models/bit/modeling_bit.py b/src/transformers/models/bit/modeling_bit.py index ec778380b6ba..1e491f06eae6 100644 --- a/src/transformers/models/bit/modeling_bit.py +++ b/src/transformers/models/bit/modeling_bit.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/blenderbot/modeling_blenderbot.py b/src/transformers/models/blenderbot/modeling_blenderbot.py index 56561612bac2..3e25fad20d31 100755 --- a/src/transformers/models/blenderbot/modeling_blenderbot.py +++ b/src/transformers/models/blenderbot/modeling_blenderbot.py @@ -20,7 +20,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py b/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py index 556fbeb4d0cb..e0e404f27cf8 100755 --- a/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py +++ b/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/blip/modeling_blip.py b/src/transformers/models/blip/modeling_blip.py index c3b5821601fb..f979518e9e11 100644 --- a/src/transformers/models/blip/modeling_blip.py +++ b/src/transformers/models/blip/modeling_blip.py @@ -19,7 +19,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn.functional import normalize diff --git a/src/transformers/models/blip/modeling_blip_text.py b/src/transformers/models/blip/modeling_blip_text.py index 0eb140685fda..6f1f58c75334 100644 --- a/src/transformers/models/blip/modeling_blip_text.py +++ b/src/transformers/models/blip/modeling_blip_text.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, device, nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/blip_2/modeling_blip_2.py b/src/transformers/models/blip_2/modeling_blip_2.py index 6a488fba8f5a..b552df47f2fc 100644 --- a/src/transformers/models/blip_2/modeling_blip_2.py +++ b/src/transformers/models/blip_2/modeling_blip_2.py @@ -20,7 +20,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/bloom/modeling_bloom.py b/src/transformers/models/bloom/modeling_bloom.py index 605ae4f59b63..6fde63e03b4d 100644 --- a/src/transformers/models/bloom/modeling_bloom.py +++ b/src/transformers/models/bloom/modeling_bloom.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, LayerNorm, MSELoss from torch.nn import functional as F diff --git a/src/transformers/models/bridgetower/modeling_bridgetower.py b/src/transformers/models/bridgetower/modeling_bridgetower.py index 2c798fcf4772..59c5be00c316 100644 --- a/src/transformers/models/bridgetower/modeling_bridgetower.py +++ b/src/transformers/models/bridgetower/modeling_bridgetower.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/bros/modeling_bros.py b/src/transformers/models/bros/modeling_bros.py index f12b47081d6a..d01a4c5a1c6d 100755 --- a/src/transformers/models/bros/modeling_bros.py +++ b/src/transformers/models/bros/modeling_bros.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/camembert/modeling_camembert.py b/src/transformers/models/camembert/modeling_camembert.py index f566bab0b8ed..3a07402f739a 100644 --- a/src/transformers/models/camembert/modeling_camembert.py +++ b/src/transformers/models/camembert/modeling_camembert.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/canine/modeling_canine.py b/src/transformers/models/canine/modeling_canine.py index 1f83c15b4f96..585961180f9e 100644 --- a/src/transformers/models/canine/modeling_canine.py +++ b/src/transformers/models/canine/modeling_canine.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/chameleon/modeling_chameleon.py b/src/transformers/models/chameleon/modeling_chameleon.py index 6b2ab15bbb9e..033b8ecd7c63 100644 --- a/src/transformers/models/chameleon/modeling_chameleon.py +++ b/src/transformers/models/chameleon/modeling_chameleon.py @@ -19,7 +19,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/chinese_clip/modeling_chinese_clip.py b/src/transformers/models/chinese_clip/modeling_chinese_clip.py index c10d0c855ce1..a0b461ab3ed3 100644 --- a/src/transformers/models/chinese_clip/modeling_chinese_clip.py +++ b/src/transformers/models/chinese_clip/modeling_chinese_clip.py @@ -18,7 +18,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/clipseg/modeling_clipseg.py b/src/transformers/models/clipseg/modeling_clipseg.py index 5938aebd1ff5..3db986aa040f 100644 --- a/src/transformers/models/clipseg/modeling_clipseg.py +++ b/src/transformers/models/clipseg/modeling_clipseg.py @@ -20,7 +20,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/clvp/modeling_clvp.py b/src/transformers/models/clvp/modeling_clvp.py index acef62d5da21..552434b5bb22 100644 --- a/src/transformers/models/clvp/modeling_clvp.py +++ b/src/transformers/models/clvp/modeling_clvp.py @@ -21,7 +21,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/codegen/modeling_codegen.py b/src/transformers/models/codegen/modeling_codegen.py index 6ccb502766cb..887b400b4799 100644 --- a/src/transformers/models/codegen/modeling_codegen.py +++ b/src/transformers/models/codegen/modeling_codegen.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/cohere/modular_cohere.py b/src/transformers/models/cohere/modular_cohere.py index 62b6b1988daa..daa12a15ed26 100644 --- a/src/transformers/models/cohere/modular_cohere.py +++ b/src/transformers/models/cohere/modular_cohere.py @@ -25,7 +25,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...cache_utils import Cache diff --git a/src/transformers/models/convbert/modeling_convbert.py b/src/transformers/models/convbert/modeling_convbert.py index 130cf183849e..080b93fa92a6 100755 --- a/src/transformers/models/convbert/modeling_convbert.py +++ b/src/transformers/models/convbert/modeling_convbert.py @@ -20,7 +20,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/convnext/modeling_convnext.py b/src/transformers/models/convnext/modeling_convnext.py index b61d18ed7294..3120c140d2ed 100755 --- a/src/transformers/models/convnext/modeling_convnext.py +++ b/src/transformers/models/convnext/modeling_convnext.py @@ -17,7 +17,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/convnextv2/modeling_convnextv2.py b/src/transformers/models/convnextv2/modeling_convnextv2.py index a5fe43f84c8b..bfa5338f5e86 100644 --- a/src/transformers/models/convnextv2/modeling_convnextv2.py +++ b/src/transformers/models/convnextv2/modeling_convnextv2.py @@ -17,7 +17,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/cpmant/modeling_cpmant.py b/src/transformers/models/cpmant/modeling_cpmant.py index 8eb5bc4d1968..1930cc0e8793 100755 --- a/src/transformers/models/cpmant/modeling_cpmant.py +++ b/src/transformers/models/cpmant/modeling_cpmant.py @@ -19,7 +19,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/cvt/modeling_cvt.py b/src/transformers/models/cvt/modeling_cvt.py index 85e2bde325e2..9d935ee84893 100644 --- a/src/transformers/models/cvt/modeling_cvt.py +++ b/src/transformers/models/cvt/modeling_cvt.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/data2vec/modeling_data2vec_text.py b/src/transformers/models/data2vec/modeling_data2vec_text.py index 9d52c9953307..f866dd9144a6 100644 --- a/src/transformers/models/data2vec/modeling_data2vec_text.py +++ b/src/transformers/models/data2vec/modeling_data2vec_text.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/data2vec/modeling_data2vec_vision.py b/src/transformers/models/data2vec/modeling_data2vec_vision.py index 6d76852122ac..f214f8eb6a0b 100644 --- a/src/transformers/models/data2vec/modeling_data2vec_vision.py +++ b/src/transformers/models/data2vec/modeling_data2vec_vision.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/dbrx/modeling_dbrx.py b/src/transformers/models/dbrx/modeling_dbrx.py index ab3ddce51310..5f3a423213cb 100644 --- a/src/transformers/models/dbrx/modeling_dbrx.py +++ b/src/transformers/models/dbrx/modeling_dbrx.py @@ -18,7 +18,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/deberta/modeling_deberta.py b/src/transformers/models/deberta/modeling_deberta.py index 0e298f52297a..461572b47677 100644 --- a/src/transformers/models/deberta/modeling_deberta.py +++ b/src/transformers/models/deberta/modeling_deberta.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deberta_v2/modeling_deberta_v2.py b/src/transformers/models/deberta_v2/modeling_deberta_v2.py index 047d4b3acd25..9d06f00c0ce6 100644 --- a/src/transformers/models/deberta_v2/modeling_deberta_v2.py +++ b/src/transformers/models/deberta_v2/modeling_deberta_v2.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, LayerNorm, MSELoss diff --git a/src/transformers/models/decision_transformer/modeling_decision_transformer.py b/src/transformers/models/decision_transformer/modeling_decision_transformer.py index 3ac0ab038219..f9c68fcbdeae 100755 --- a/src/transformers/models/decision_transformer/modeling_decision_transformer.py +++ b/src/transformers/models/decision_transformer/modeling_decision_transformer.py @@ -20,7 +20,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/deepseek_v3/modular_deepseek_v3.py b/src/transformers/models/deepseek_v3/modular_deepseek_v3.py index 38cc8dbb5ea1..fc3dc0c4ce3b 100644 --- a/src/transformers/models/deepseek_v3/modular_deepseek_v3.py +++ b/src/transformers/models/deepseek_v3/modular_deepseek_v3.py @@ -3,7 +3,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/deit/modeling_deit.py b/src/transformers/models/deit/modeling_deit.py index 6a6be311137d..4015dcbe0bc3 100644 --- a/src/transformers/models/deit/modeling_deit.py +++ b/src/transformers/models/deit/modeling_deit.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/deprecated/efficientformer/modeling_efficientformer.py b/src/transformers/models/deprecated/efficientformer/modeling_efficientformer.py index 3d918e7f5720..d35d3e82c007 100644 --- a/src/transformers/models/deprecated/efficientformer/modeling_efficientformer.py +++ b/src/transformers/models/deprecated/efficientformer/modeling_efficientformer.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ....activations import ACT2FN diff --git a/src/transformers/models/deprecated/ernie_m/modeling_ernie_m.py b/src/transformers/models/deprecated/ernie_m/modeling_ernie_m.py index 90f215157b7e..f0e97c132d09 100755 --- a/src/transformers/models/deprecated/ernie_m/modeling_ernie_m.py +++ b/src/transformers/models/deprecated/ernie_m/modeling_ernie_m.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn, tensor from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deprecated/mctct/modeling_mctct.py b/src/transformers/models/deprecated/mctct/modeling_mctct.py index 3c9d259e8215..253b09c1c43c 100755 --- a/src/transformers/models/deprecated/mctct/modeling_mctct.py +++ b/src/transformers/models/deprecated/mctct/modeling_mctct.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ....activations import ACT2FN diff --git a/src/transformers/models/deprecated/mega/modeling_mega.py b/src/transformers/models/deprecated/mega/modeling_mega.py index c6edc57f8cf6..c237afee9a33 100644 --- a/src/transformers/models/deprecated/mega/modeling_mega.py +++ b/src/transformers/models/deprecated/mega/modeling_mega.py @@ -19,7 +19,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deprecated/nat/modeling_nat.py b/src/transformers/models/deprecated/nat/modeling_nat.py index d463e2bc89cc..a619cdb11225 100644 --- a/src/transformers/models/deprecated/nat/modeling_nat.py +++ b/src/transformers/models/deprecated/nat/modeling_nat.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ....activations import ACT2FN diff --git a/src/transformers/models/deprecated/nezha/modeling_nezha.py b/src/transformers/models/deprecated/nezha/modeling_nezha.py index 3dd67c22d72e..ddfecac9f506 100644 --- a/src/transformers/models/deprecated/nezha/modeling_nezha.py +++ b/src/transformers/models/deprecated/nezha/modeling_nezha.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deprecated/open_llama/modeling_open_llama.py b/src/transformers/models/deprecated/open_llama/modeling_open_llama.py index 4ce63feceb74..5e182e0f813f 100644 --- a/src/transformers/models/deprecated/open_llama/modeling_open_llama.py +++ b/src/transformers/models/deprecated/open_llama/modeling_open_llama.py @@ -23,7 +23,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deprecated/qdqbert/modeling_qdqbert.py b/src/transformers/models/deprecated/qdqbert/modeling_qdqbert.py index 74bad366e8d5..f92bc07a8bfb 100755 --- a/src/transformers/models/deprecated/qdqbert/modeling_qdqbert.py +++ b/src/transformers/models/deprecated/qdqbert/modeling_qdqbert.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deprecated/trajectory_transformer/modeling_trajectory_transformer.py b/src/transformers/models/deprecated/trajectory_transformer/modeling_trajectory_transformer.py index dcacffabc8b9..b6ae410c1474 100644 --- a/src/transformers/models/deprecated/trajectory_transformer/modeling_trajectory_transformer.py +++ b/src/transformers/models/deprecated/trajectory_transformer/modeling_trajectory_transformer.py @@ -21,7 +21,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import functional as F diff --git a/src/transformers/models/deprecated/tvlt/modeling_tvlt.py b/src/transformers/models/deprecated/tvlt/modeling_tvlt.py index 5f34083ac2ff..2b21df928ff3 100644 --- a/src/transformers/models/deprecated/tvlt/modeling_tvlt.py +++ b/src/transformers/models/deprecated/tvlt/modeling_tvlt.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/deprecated/van/modeling_van.py b/src/transformers/models/deprecated/van/modeling_van.py index 025234e4e71f..c0fc0bc1a637 100644 --- a/src/transformers/models/deprecated/van/modeling_van.py +++ b/src/transformers/models/deprecated/van/modeling_van.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ....activations import ACT2FN diff --git a/src/transformers/models/deprecated/vit_hybrid/modeling_vit_hybrid.py b/src/transformers/models/deprecated/vit_hybrid/modeling_vit_hybrid.py index 2d92655cc3e4..86b1594a20c9 100644 --- a/src/transformers/models/deprecated/vit_hybrid/modeling_vit_hybrid.py +++ b/src/transformers/models/deprecated/vit_hybrid/modeling_vit_hybrid.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ....activations import ACT2FN diff --git a/src/transformers/models/deprecated/xlm_prophetnet/modeling_xlm_prophetnet.py b/src/transformers/models/deprecated/xlm_prophetnet/modeling_xlm_prophetnet.py index e85660223b58..3c4dc3de8393 100644 --- a/src/transformers/models/deprecated/xlm_prophetnet/modeling_xlm_prophetnet.py +++ b/src/transformers/models/deprecated/xlm_prophetnet/modeling_xlm_prophetnet.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from torch.nn import LayerNorm diff --git a/src/transformers/models/depth_anything/modeling_depth_anything.py b/src/transformers/models/depth_anything/modeling_depth_anything.py index 06a3a8d300b8..bc7d74131204 100644 --- a/src/transformers/models/depth_anything/modeling_depth_anything.py +++ b/src/transformers/models/depth_anything/modeling_depth_anything.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...modeling_outputs import DepthEstimatorOutput diff --git a/src/transformers/models/dinat/modeling_dinat.py b/src/transformers/models/dinat/modeling_dinat.py index 384bdee49d35..4b7ec37b0ea8 100644 --- a/src/transformers/models/dinat/modeling_dinat.py +++ b/src/transformers/models/dinat/modeling_dinat.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/dinov2/modeling_dinov2.py b/src/transformers/models/dinov2/modeling_dinov2.py index 96a051327e01..0a9a2cba1da7 100644 --- a/src/transformers/models/dinov2/modeling_dinov2.py +++ b/src/transformers/models/dinov2/modeling_dinov2.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/dinov2_with_registers/modular_dinov2_with_registers.py b/src/transformers/models/dinov2_with_registers/modular_dinov2_with_registers.py index c23e523e3434..686528002b09 100644 --- a/src/transformers/models/dinov2_with_registers/modular_dinov2_with_registers.py +++ b/src/transformers/models/dinov2_with_registers/modular_dinov2_with_registers.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ....transformers.models.dinov2.modeling_dinov2 import ( diff --git a/src/transformers/models/dinov3_convnext/modeling_dinov3_convnext.py b/src/transformers/models/dinov3_convnext/modeling_dinov3_convnext.py index 2318faf14824..df2ef491192c 100644 --- a/src/transformers/models/dinov3_convnext/modeling_dinov3_convnext.py +++ b/src/transformers/models/dinov3_convnext/modeling_dinov3_convnext.py @@ -18,7 +18,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/dinov3_vit/modular_dinov3_vit.py b/src/transformers/models/dinov3_vit/modular_dinov3_vit.py index f4a1e69beaac..0515a1a1e0bf 100644 --- a/src/transformers/models/dinov3_vit/modular_dinov3_vit.py +++ b/src/transformers/models/dinov3_vit/modular_dinov3_vit.py @@ -19,7 +19,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from transformers.models.arcee.modeling_arcee import ArceeMLP diff --git a/src/transformers/models/donut/modeling_donut_swin.py b/src/transformers/models/donut/modeling_donut_swin.py index 882fd72c508e..c5736b16183b 100644 --- a/src/transformers/models/donut/modeling_donut_swin.py +++ b/src/transformers/models/donut/modeling_donut_swin.py @@ -23,7 +23,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/dpt/modeling_dpt.py b/src/transformers/models/dpt/modeling_dpt.py index f797f53aa4b6..363fce92f897 100755 --- a/src/transformers/models/dpt/modeling_dpt.py +++ b/src/transformers/models/dpt/modeling_dpt.py @@ -24,7 +24,6 @@ from typing import Callable, Optional import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/efficientnet/modeling_efficientnet.py b/src/transformers/models/efficientnet/modeling_efficientnet.py index 70ec3914f7de..a263ff20760c 100644 --- a/src/transformers/models/efficientnet/modeling_efficientnet.py +++ b/src/transformers/models/efficientnet/modeling_efficientnet.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/electra/modeling_electra.py b/src/transformers/models/electra/modeling_electra.py index 9327bc0fdf26..a10b0b658337 100644 --- a/src/transformers/models/electra/modeling_electra.py +++ b/src/transformers/models/electra/modeling_electra.py @@ -20,7 +20,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/emu3/modular_emu3.py b/src/transformers/models/emu3/modular_emu3.py index 5dd8d02f61aa..32599727b24c 100644 --- a/src/transformers/models/emu3/modular_emu3.py +++ b/src/transformers/models/emu3/modular_emu3.py @@ -21,7 +21,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from ...cache_utils import Cache from ...generation import GenerationMixin diff --git a/src/transformers/models/ernie/modeling_ernie.py b/src/transformers/models/ernie/modeling_ernie.py index a5ed4a3f5328..7cbce6b2d20b 100644 --- a/src/transformers/models/ernie/modeling_ernie.py +++ b/src/transformers/models/ernie/modeling_ernie.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/esm/modeling_esm.py b/src/transformers/models/esm/modeling_esm.py index 5db366aa6197..ddcf460f01ee 100755 --- a/src/transformers/models/esm/modeling_esm.py +++ b/src/transformers/models/esm/modeling_esm.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/evolla/modular_evolla.py b/src/transformers/models/evolla/modular_evolla.py index a58a3e7b7341..18a50e9abfae 100644 --- a/src/transformers/models/evolla/modular_evolla.py +++ b/src/transformers/models/evolla/modular_evolla.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...cache_utils import Cache, DynamicCache diff --git a/src/transformers/models/falcon/modeling_falcon.py b/src/transformers/models/falcon/modeling_falcon.py index 5ec1792e4581..26dc56e41480 100644 --- a/src/transformers/models/falcon/modeling_falcon.py +++ b/src/transformers/models/falcon/modeling_falcon.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, LayerNorm, MSELoss from torch.nn import functional as F diff --git a/src/transformers/models/falcon_h1/modular_falcon_h1.py b/src/transformers/models/falcon_h1/modular_falcon_h1.py index c81e8967bcf2..24eb98ccd1ed 100644 --- a/src/transformers/models/falcon_h1/modular_falcon_h1.py +++ b/src/transformers/models/falcon_h1/modular_falcon_h1.py @@ -23,7 +23,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from transformers.activations import ACT2FN diff --git a/src/transformers/models/falcon_mamba/modular_falcon_mamba.py b/src/transformers/models/falcon_mamba/modular_falcon_mamba.py index cfe2ec49a992..6df2be3a2652 100644 --- a/src/transformers/models/falcon_mamba/modular_falcon_mamba.py +++ b/src/transformers/models/falcon_mamba/modular_falcon_mamba.py @@ -17,7 +17,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import nn from ...utils import auto_docstring, logging diff --git a/src/transformers/models/flava/modeling_flava.py b/src/transformers/models/flava/modeling_flava.py index cefaa6f95f68..266c3e96af5a 100644 --- a/src/transformers/models/flava/modeling_flava.py +++ b/src/transformers/models/flava/modeling_flava.py @@ -21,7 +21,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/fnet/modeling_fnet.py b/src/transformers/models/fnet/modeling_fnet.py index 1cb0e764b2a1..2ad09a3b268b 100755 --- a/src/transformers/models/fnet/modeling_fnet.py +++ b/src/transformers/models/fnet/modeling_fnet.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/focalnet/modeling_focalnet.py b/src/transformers/models/focalnet/modeling_focalnet.py index e56ada740e22..ed31b5deb527 100644 --- a/src/transformers/models/focalnet/modeling_focalnet.py +++ b/src/transformers/models/focalnet/modeling_focalnet.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/fuyu/modeling_fuyu.py b/src/transformers/models/fuyu/modeling_fuyu.py index d5edfadc3ffc..2095e9877c2c 100644 --- a/src/transformers/models/fuyu/modeling_fuyu.py +++ b/src/transformers/models/fuyu/modeling_fuyu.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...cache_utils import Cache diff --git a/src/transformers/models/gemma2/modular_gemma2.py b/src/transformers/models/gemma2/modular_gemma2.py index 47d612de5a4b..c7e34e4abed4 100644 --- a/src/transformers/models/gemma2/modular_gemma2.py +++ b/src/transformers/models/gemma2/modular_gemma2.py @@ -17,7 +17,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from ...activations import ACT2FN from ...cache_utils import Cache, DynamicCache diff --git a/src/transformers/models/gemma3/modular_gemma3.py b/src/transformers/models/gemma3/modular_gemma3.py index f0658f9825f8..a6e8b93e822f 100644 --- a/src/transformers/models/gemma3/modular_gemma3.py +++ b/src/transformers/models/gemma3/modular_gemma3.py @@ -19,7 +19,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from ...cache_utils import Cache, DynamicCache from ...configuration_utils import PretrainedConfig, layer_type_validation diff --git a/src/transformers/models/git/modeling_git.py b/src/transformers/models/git/modeling_git.py index 0125132718a3..4122b7a0df79 100644 --- a/src/transformers/models/git/modeling_git.py +++ b/src/transformers/models/git/modeling_git.py @@ -20,7 +20,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/glm/modular_glm.py b/src/transformers/models/glm/modular_glm.py index ec07be10fb6a..90730c0184a3 100644 --- a/src/transformers/models/glm/modular_glm.py +++ b/src/transformers/models/glm/modular_glm.py @@ -17,7 +17,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from ...utils import logging from ..llama.modeling_llama import ( diff --git a/src/transformers/models/glm4_moe/modular_glm4_moe.py b/src/transformers/models/glm4_moe/modular_glm4_moe.py index bc07483c7f22..20144c8ffc40 100644 --- a/src/transformers/models/glm4_moe/modular_glm4_moe.py +++ b/src/transformers/models/glm4_moe/modular_glm4_moe.py @@ -17,7 +17,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import nn from ...configuration_utils import PretrainedConfig diff --git a/src/transformers/models/glpn/modeling_glpn.py b/src/transformers/models/glpn/modeling_glpn.py index 65e7b9b2654d..e326750743a1 100755 --- a/src/transformers/models/glpn/modeling_glpn.py +++ b/src/transformers/models/glpn/modeling_glpn.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/gpt2/modeling_gpt2.py b/src/transformers/models/gpt2/modeling_gpt2.py index c34755b8b440..ae0786179464 100644 --- a/src/transformers/models/gpt2/modeling_gpt2.py +++ b/src/transformers/models/gpt2/modeling_gpt2.py @@ -22,7 +22,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/gpt_bigcode/modeling_gpt_bigcode.py b/src/transformers/models/gpt_bigcode/modeling_gpt_bigcode.py index 96fb40b1d69f..6992dc642a4f 100644 --- a/src/transformers/models/gpt_bigcode/modeling_gpt_bigcode.py +++ b/src/transformers/models/gpt_bigcode/modeling_gpt_bigcode.py @@ -17,7 +17,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/gpt_neo/modeling_gpt_neo.py b/src/transformers/models/gpt_neo/modeling_gpt_neo.py index f7cf160cbb21..69d74565745a 100755 --- a/src/transformers/models/gpt_neo/modeling_gpt_neo.py +++ b/src/transformers/models/gpt_neo/modeling_gpt_neo.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/gpt_neox/modular_gpt_neox.py b/src/transformers/models/gpt_neox/modular_gpt_neox.py index 1d808304c306..532b7a607ae8 100644 --- a/src/transformers/models/gpt_neox/modular_gpt_neox.py +++ b/src/transformers/models/gpt_neox/modular_gpt_neox.py @@ -1,7 +1,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/gpt_neox_japanese/modeling_gpt_neox_japanese.py b/src/transformers/models/gpt_neox_japanese/modeling_gpt_neox_japanese.py index e25548d90f0c..70399f376c75 100755 --- a/src/transformers/models/gpt_neox_japanese/modeling_gpt_neox_japanese.py +++ b/src/transformers/models/gpt_neox_japanese/modeling_gpt_neox_japanese.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/gptj/modeling_gptj.py b/src/transformers/models/gptj/modeling_gptj.py index 56256df7d582..cb6a4f579c52 100644 --- a/src/transformers/models/gptj/modeling_gptj.py +++ b/src/transformers/models/gptj/modeling_gptj.py @@ -19,7 +19,6 @@ import torch import torch.fx -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/granite/modular_granite.py b/src/transformers/models/granite/modular_granite.py index 1b90609e5460..37e1955fcb09 100644 --- a/src/transformers/models/granite/modular_granite.py +++ b/src/transformers/models/granite/modular_granite.py @@ -16,7 +16,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...cache_utils import Cache, DynamicCache diff --git a/src/transformers/models/groupvit/modeling_groupvit.py b/src/transformers/models/groupvit/modeling_groupvit.py index 9ea10095eec8..775ebd286f0a 100644 --- a/src/transformers/models/groupvit/modeling_groupvit.py +++ b/src/transformers/models/groupvit/modeling_groupvit.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/helium/modular_helium.py b/src/transformers/models/helium/modular_helium.py index fe53f7820abb..6c2538d438f9 100644 --- a/src/transformers/models/helium/modular_helium.py +++ b/src/transformers/models/helium/modular_helium.py @@ -18,7 +18,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from ...utils import logging from ..gemma.modeling_gemma import GemmaForCausalLM, GemmaForSequenceClassification, GemmaForTokenClassification diff --git a/src/transformers/models/hiera/modeling_hiera.py b/src/transformers/models/hiera/modeling_hiera.py index bfef87618156..0c084f0f836e 100644 --- a/src/transformers/models/hiera/modeling_hiera.py +++ b/src/transformers/models/hiera/modeling_hiera.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/hunyuan_v1_dense/modular_hunyuan_v1_dense.py b/src/transformers/models/hunyuan_v1_dense/modular_hunyuan_v1_dense.py index c79ccc6a616d..d527abc08f93 100644 --- a/src/transformers/models/hunyuan_v1_dense/modular_hunyuan_v1_dense.py +++ b/src/transformers/models/hunyuan_v1_dense/modular_hunyuan_v1_dense.py @@ -17,7 +17,6 @@ from typing import Callable, Optional import torch -import torch.utils.checkpoint from torch import nn from transformers.cache_utils import Cache diff --git a/src/transformers/models/hunyuan_v1_moe/modular_hunyuan_v1_moe.py b/src/transformers/models/hunyuan_v1_moe/modular_hunyuan_v1_moe.py index 645c54ae73af..a72d6268fe70 100644 --- a/src/transformers/models/hunyuan_v1_moe/modular_hunyuan_v1_moe.py +++ b/src/transformers/models/hunyuan_v1_moe/modular_hunyuan_v1_moe.py @@ -18,7 +18,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from transformers.cache_utils import Cache diff --git a/src/transformers/models/ibert/modeling_ibert.py b/src/transformers/models/ibert/modeling_ibert.py index 6b960148ca9b..57b3df2f570b 100644 --- a/src/transformers/models/ibert/modeling_ibert.py +++ b/src/transformers/models/ibert/modeling_ibert.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/idefics/modeling_idefics.py b/src/transformers/models/idefics/modeling_idefics.py index c80cb2e88bdc..f2fb135a4f4e 100644 --- a/src/transformers/models/idefics/modeling_idefics.py +++ b/src/transformers/models/idefics/modeling_idefics.py @@ -24,7 +24,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/idefics/vision.py b/src/transformers/models/idefics/vision.py index 72521761d9d1..f6143064835e 100644 --- a/src/transformers/models/idefics/vision.py +++ b/src/transformers/models/idefics/vision.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/idefics2/modeling_idefics2.py b/src/transformers/models/idefics2/modeling_idefics2.py index 264d3fc831ab..3aab4b01977f 100644 --- a/src/transformers/models/idefics2/modeling_idefics2.py +++ b/src/transformers/models/idefics2/modeling_idefics2.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/idefics3/modeling_idefics3.py b/src/transformers/models/idefics3/modeling_idefics3.py index 9d726f814465..32c75c1f3667 100644 --- a/src/transformers/models/idefics3/modeling_idefics3.py +++ b/src/transformers/models/idefics3/modeling_idefics3.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/imagegpt/modeling_imagegpt.py b/src/transformers/models/imagegpt/modeling_imagegpt.py index 6a424727d8e5..a962141e4479 100755 --- a/src/transformers/models/imagegpt/modeling_imagegpt.py +++ b/src/transformers/models/imagegpt/modeling_imagegpt.py @@ -19,7 +19,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/instructblip/modeling_instructblip.py b/src/transformers/models/instructblip/modeling_instructblip.py index af039a508c5d..20c0def10fd1 100644 --- a/src/transformers/models/instructblip/modeling_instructblip.py +++ b/src/transformers/models/instructblip/modeling_instructblip.py @@ -19,7 +19,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/instructblipvideo/modular_instructblipvideo.py b/src/transformers/models/instructblipvideo/modular_instructblipvideo.py index ff30263700cf..5619c2e79b9a 100644 --- a/src/transformers/models/instructblipvideo/modular_instructblipvideo.py +++ b/src/transformers/models/instructblipvideo/modular_instructblipvideo.py @@ -16,7 +16,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from transformers.models.instructblip.configuration_instructblip import ( InstructBlipQFormerConfig, diff --git a/src/transformers/models/internvl/modular_internvl.py b/src/transformers/models/internvl/modular_internvl.py index bcef3a2ccbb0..ac7a4f516c06 100644 --- a/src/transformers/models/internvl/modular_internvl.py +++ b/src/transformers/models/internvl/modular_internvl.py @@ -20,7 +20,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from ...activations import ACT2FN from ...cache_utils import Cache diff --git a/src/transformers/models/jamba/modeling_jamba.py b/src/transformers/models/jamba/modeling_jamba.py index 17246d6f1b2e..c8ddeb970e26 100755 --- a/src/transformers/models/jamba/modeling_jamba.py +++ b/src/transformers/models/jamba/modeling_jamba.py @@ -24,7 +24,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/janus/modular_janus.py b/src/transformers/models/janus/modular_janus.py index 261e994262aa..7f75147a5b1e 100644 --- a/src/transformers/models/janus/modular_janus.py +++ b/src/transformers/models/janus/modular_janus.py @@ -70,7 +70,7 @@ import torch import torch.nn as nn import torch.nn.functional as F - import torch.utils.checkpoint + if is_vision_available(): import PIL diff --git a/src/transformers/models/jetmoe/modeling_jetmoe.py b/src/transformers/models/jetmoe/modeling_jetmoe.py index 896426a5e320..0ca0a9a43669 100644 --- a/src/transformers/models/jetmoe/modeling_jetmoe.py +++ b/src/transformers/models/jetmoe/modeling_jetmoe.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import functional as F diff --git a/src/transformers/models/kosmos2/modeling_kosmos2.py b/src/transformers/models/kosmos2/modeling_kosmos2.py index 0372ec92a6ee..76acda9f0de9 100644 --- a/src/transformers/models/kosmos2/modeling_kosmos2.py +++ b/src/transformers/models/kosmos2/modeling_kosmos2.py @@ -19,7 +19,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/kosmos2_5/modeling_kosmos2_5.py b/src/transformers/models/kosmos2_5/modeling_kosmos2_5.py index 1bb70fd5093d..8f9fbd706b32 100644 --- a/src/transformers/models/kosmos2_5/modeling_kosmos2_5.py +++ b/src/transformers/models/kosmos2_5/modeling_kosmos2_5.py @@ -19,7 +19,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/layoutlm/modeling_layoutlm.py b/src/transformers/models/layoutlm/modeling_layoutlm.py index b3b79ef99d38..9e71eb7d8fb9 100644 --- a/src/transformers/models/layoutlm/modeling_layoutlm.py +++ b/src/transformers/models/layoutlm/modeling_layoutlm.py @@ -17,7 +17,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/layoutlmv2/modeling_layoutlmv2.py b/src/transformers/models/layoutlmv2/modeling_layoutlmv2.py index 11d8127ef6c2..3f444fbb6b28 100755 --- a/src/transformers/models/layoutlmv2/modeling_layoutlmv2.py +++ b/src/transformers/models/layoutlmv2/modeling_layoutlmv2.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/layoutlmv3/modeling_layoutlmv3.py b/src/transformers/models/layoutlmv3/modeling_layoutlmv3.py index bd8b525bb427..73bf26b0dfbe 100644 --- a/src/transformers/models/layoutlmv3/modeling_layoutlmv3.py +++ b/src/transformers/models/layoutlmv3/modeling_layoutlmv3.py @@ -21,7 +21,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss from ...activations import ACT2FN diff --git a/src/transformers/models/led/modeling_led.py b/src/transformers/models/led/modeling_led.py index e34a261df552..26d1321842e6 100755 --- a/src/transformers/models/led/modeling_led.py +++ b/src/transformers/models/led/modeling_led.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/levit/modeling_levit.py b/src/transformers/models/levit/modeling_levit.py index a72f5604825f..3deca07e2400 100644 --- a/src/transformers/models/levit/modeling_levit.py +++ b/src/transformers/models/levit/modeling_levit.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...modeling_outputs import ( diff --git a/src/transformers/models/lilt/modeling_lilt.py b/src/transformers/models/lilt/modeling_lilt.py index c3bcbf31f035..bb00d16c3965 100644 --- a/src/transformers/models/lilt/modeling_lilt.py +++ b/src/transformers/models/lilt/modeling_lilt.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/llava/modeling_llava.py b/src/transformers/models/llava/modeling_llava.py index 9a116dac4d23..bc0bb0df7c7b 100644 --- a/src/transformers/models/llava/modeling_llava.py +++ b/src/transformers/models/llava/modeling_llava.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/longformer/modeling_longformer.py b/src/transformers/models/longformer/modeling_longformer.py index f181217cd101..cdc708924967 100755 --- a/src/transformers/models/longformer/modeling_longformer.py +++ b/src/transformers/models/longformer/modeling_longformer.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/luke/modeling_luke.py b/src/transformers/models/luke/modeling_luke.py index ed1f2084c9e8..e78197beeb57 100644 --- a/src/transformers/models/luke/modeling_luke.py +++ b/src/transformers/models/luke/modeling_luke.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/mamba/modeling_mamba.py b/src/transformers/models/mamba/modeling_mamba.py index 4a53c47c8b4a..487cb3d19306 100644 --- a/src/transformers/models/mamba/modeling_mamba.py +++ b/src/transformers/models/mamba/modeling_mamba.py @@ -19,7 +19,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/mamba2/modeling_mamba2.py b/src/transformers/models/mamba2/modeling_mamba2.py index 738c5376c33e..a423c5b42fbd 100644 --- a/src/transformers/models/mamba2/modeling_mamba2.py +++ b/src/transformers/models/mamba2/modeling_mamba2.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/marian/modeling_marian.py b/src/transformers/models/marian/modeling_marian.py index 342e622321a8..f5f567346412 100755 --- a/src/transformers/models/marian/modeling_marian.py +++ b/src/transformers/models/marian/modeling_marian.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/markuplm/modeling_markuplm.py b/src/transformers/models/markuplm/modeling_markuplm.py index aeb817be7060..78fbf8f215aa 100755 --- a/src/transformers/models/markuplm/modeling_markuplm.py +++ b/src/transformers/models/markuplm/modeling_markuplm.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/mbart/modeling_mbart.py b/src/transformers/models/mbart/modeling_mbart.py index 21c54b6de60e..55fad55a87ae 100755 --- a/src/transformers/models/mbart/modeling_mbart.py +++ b/src/transformers/models/mbart/modeling_mbart.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/megatron_bert/modeling_megatron_bert.py b/src/transformers/models/megatron_bert/modeling_megatron_bert.py index 4987bd15dffd..a75c0f575aca 100755 --- a/src/transformers/models/megatron_bert/modeling_megatron_bert.py +++ b/src/transformers/models/megatron_bert/modeling_megatron_bert.py @@ -22,7 +22,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/mgp_str/modeling_mgp_str.py b/src/transformers/models/mgp_str/modeling_mgp_str.py index 8f65375a7895..be7cf08b14ab 100644 --- a/src/transformers/models/mgp_str/modeling_mgp_str.py +++ b/src/transformers/models/mgp_str/modeling_mgp_str.py @@ -20,7 +20,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...modeling_outputs import BaseModelOutput diff --git a/src/transformers/models/mimi/modeling_mimi.py b/src/transformers/models/mimi/modeling_mimi.py index 64537d5fcd94..f22cad968247 100644 --- a/src/transformers/models/mimi/modeling_mimi.py +++ b/src/transformers/models/mimi/modeling_mimi.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/mixtral/modular_mixtral.py b/src/transformers/models/mixtral/modular_mixtral.py index d897824c4cff..744f8c1321dc 100644 --- a/src/transformers/models/mixtral/modular_mixtral.py +++ b/src/transformers/models/mixtral/modular_mixtral.py @@ -23,7 +23,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/mllama/modeling_mllama.py b/src/transformers/models/mllama/modeling_mllama.py index e30ac5a8fa64..eb6cbee777d6 100644 --- a/src/transformers/models/mllama/modeling_mllama.py +++ b/src/transformers/models/mllama/modeling_mllama.py @@ -19,7 +19,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/mobilevit/modeling_mobilevit.py b/src/transformers/models/mobilevit/modeling_mobilevit.py index 10fe620f7c0d..415c33a7cb85 100755 --- a/src/transformers/models/mobilevit/modeling_mobilevit.py +++ b/src/transformers/models/mobilevit/modeling_mobilevit.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/mobilevitv2/modeling_mobilevitv2.py b/src/transformers/models/mobilevitv2/modeling_mobilevitv2.py index 291ce6136a54..4e0e972a648a 100644 --- a/src/transformers/models/mobilevitv2/modeling_mobilevitv2.py +++ b/src/transformers/models/mobilevitv2/modeling_mobilevitv2.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/modernbert/modular_modernbert.py b/src/transformers/models/modernbert/modular_modernbert.py index 394fdce4fe80..276a754cc101 100644 --- a/src/transformers/models/modernbert/modular_modernbert.py +++ b/src/transformers/models/modernbert/modular_modernbert.py @@ -20,7 +20,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/moshi/modeling_moshi.py b/src/transformers/models/moshi/modeling_moshi.py index 503177f95b4a..7546fc90e542 100644 --- a/src/transformers/models/moshi/modeling_moshi.py +++ b/src/transformers/models/moshi/modeling_moshi.py @@ -20,7 +20,6 @@ import torch import torch.nn as nn -import torch.utils.checkpoint from torch.nn import CrossEntropyLoss from ...activations import ACT2FN diff --git a/src/transformers/models/mpt/modeling_mpt.py b/src/transformers/models/mpt/modeling_mpt.py index 57b875432758..c7bf0a795d42 100644 --- a/src/transformers/models/mpt/modeling_mpt.py +++ b/src/transformers/models/mpt/modeling_mpt.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, LayerNorm, MSELoss from torch.nn import functional as F diff --git a/src/transformers/models/mra/modeling_mra.py b/src/transformers/models/mra/modeling_mra.py index 3a37712e8580..86bee4d09b5a 100644 --- a/src/transformers/models/mra/modeling_mra.py +++ b/src/transformers/models/mra/modeling_mra.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss from torch.utils.cpp_extension import load diff --git a/src/transformers/models/mvp/modeling_mvp.py b/src/transformers/models/mvp/modeling_mvp.py index 22af2b5a74de..6838f209cb4e 100644 --- a/src/transformers/models/mvp/modeling_mvp.py +++ b/src/transformers/models/mvp/modeling_mvp.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/nemotron/modeling_nemotron.py b/src/transformers/models/nemotron/modeling_nemotron.py index 35b1aedb71f8..b98f0a5ef2ac 100644 --- a/src/transformers/models/nemotron/modeling_nemotron.py +++ b/src/transformers/models/nemotron/modeling_nemotron.py @@ -20,7 +20,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import Size, Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/nystromformer/modeling_nystromformer.py b/src/transformers/models/nystromformer/modeling_nystromformer.py index 45e69b6b4693..3eb1fad24019 100755 --- a/src/transformers/models/nystromformer/modeling_nystromformer.py +++ b/src/transformers/models/nystromformer/modeling_nystromformer.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/olmo/modular_olmo.py b/src/transformers/models/olmo/modular_olmo.py index f54b9106345e..c89e927e4e08 100644 --- a/src/transformers/models/olmo/modular_olmo.py +++ b/src/transformers/models/olmo/modular_olmo.py @@ -3,7 +3,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from ...cache_utils import Cache from ...modeling_utils import ALL_ATTENTION_FUNCTIONS diff --git a/src/transformers/models/olmoe/modeling_olmoe.py b/src/transformers/models/olmoe/modeling_olmoe.py index 9d7d0727da9b..4070d3b2b480 100644 --- a/src/transformers/models/olmoe/modeling_olmoe.py +++ b/src/transformers/models/olmoe/modeling_olmoe.py @@ -16,7 +16,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/opt/modeling_opt.py b/src/transformers/models/opt/modeling_opt.py index a2c4bb500a65..68af2a02017b 100644 --- a/src/transformers/models/opt/modeling_opt.py +++ b/src/transformers/models/opt/modeling_opt.py @@ -17,7 +17,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/owlv2/modeling_owlv2.py b/src/transformers/models/owlv2/modeling_owlv2.py index 81c7a088df9f..715df44f01f0 100644 --- a/src/transformers/models/owlv2/modeling_owlv2.py +++ b/src/transformers/models/owlv2/modeling_owlv2.py @@ -19,7 +19,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/owlvit/modeling_owlvit.py b/src/transformers/models/owlvit/modeling_owlvit.py index 107c8a9dab2f..3971b1376d9c 100644 --- a/src/transformers/models/owlvit/modeling_owlvit.py +++ b/src/transformers/models/owlvit/modeling_owlvit.py @@ -19,7 +19,6 @@ from typing import Any, Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/paligemma/modeling_paligemma.py b/src/transformers/models/paligemma/modeling_paligemma.py index 5600af24344f..4105af3ab2e1 100644 --- a/src/transformers/models/paligemma/modeling_paligemma.py +++ b/src/transformers/models/paligemma/modeling_paligemma.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...cache_utils import Cache, StaticCache diff --git a/src/transformers/models/pegasus/modeling_pegasus.py b/src/transformers/models/pegasus/modeling_pegasus.py index 58eedc77bc3c..dc3a8005acac 100755 --- a/src/transformers/models/pegasus/modeling_pegasus.py +++ b/src/transformers/models/pegasus/modeling_pegasus.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/pegasus_x/modeling_pegasus_x.py b/src/transformers/models/pegasus_x/modeling_pegasus_x.py index 0279688c00e8..0c1ae32cabe2 100755 --- a/src/transformers/models/pegasus_x/modeling_pegasus_x.py +++ b/src/transformers/models/pegasus_x/modeling_pegasus_x.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/perceiver/modeling_perceiver.py b/src/transformers/models/perceiver/modeling_perceiver.py index 1f6b84343d00..f0e4e3e5dbe0 100755 --- a/src/transformers/models/perceiver/modeling_perceiver.py +++ b/src/transformers/models/perceiver/modeling_perceiver.py @@ -24,7 +24,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/perception_lm/modular_perception_lm.py b/src/transformers/models/perception_lm/modular_perception_lm.py index 2e748c82a7bc..2b50b8242202 100644 --- a/src/transformers/models/perception_lm/modular_perception_lm.py +++ b/src/transformers/models/perception_lm/modular_perception_lm.py @@ -18,7 +18,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...cache_utils import Cache diff --git a/src/transformers/models/persimmon/modeling_persimmon.py b/src/transformers/models/persimmon/modeling_persimmon.py index 02ec819315b7..c963bb53852a 100644 --- a/src/transformers/models/persimmon/modeling_persimmon.py +++ b/src/transformers/models/persimmon/modeling_persimmon.py @@ -22,7 +22,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/phi3/modular_phi3.py b/src/transformers/models/phi3/modular_phi3.py index a945fb21b935..d355c3792a6b 100644 --- a/src/transformers/models/phi3/modular_phi3.py +++ b/src/transformers/models/phi3/modular_phi3.py @@ -18,7 +18,6 @@ from typing import Callable, Optional import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/phimoe/modeling_phimoe.py b/src/transformers/models/phimoe/modeling_phimoe.py index d3d79847b073..711aabd4b962 100644 --- a/src/transformers/models/phimoe/modeling_phimoe.py +++ b/src/transformers/models/phimoe/modeling_phimoe.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/pix2struct/modeling_pix2struct.py b/src/transformers/models/pix2struct/modeling_pix2struct.py index d780dc534879..463fec98256f 100644 --- a/src/transformers/models/pix2struct/modeling_pix2struct.py +++ b/src/transformers/models/pix2struct/modeling_pix2struct.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/pixtral/modeling_pixtral.py b/src/transformers/models/pixtral/modeling_pixtral.py index 564c118fccb9..79bf0ee6bbda 100644 --- a/src/transformers/models/pixtral/modeling_pixtral.py +++ b/src/transformers/models/pixtral/modeling_pixtral.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/plbart/modular_plbart.py b/src/transformers/models/plbart/modular_plbart.py index 8d7f0022cfb4..29c253144557 100644 --- a/src/transformers/models/plbart/modular_plbart.py +++ b/src/transformers/models/plbart/modular_plbart.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/poolformer/modeling_poolformer.py b/src/transformers/models/poolformer/modeling_poolformer.py index 3753eb464b04..8c6dc8191630 100755 --- a/src/transformers/models/poolformer/modeling_poolformer.py +++ b/src/transformers/models/poolformer/modeling_poolformer.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/prophetnet/modeling_prophetnet.py b/src/transformers/models/prophetnet/modeling_prophetnet.py index 5e80ee4f0faa..260b0c698407 100644 --- a/src/transformers/models/prophetnet/modeling_prophetnet.py +++ b/src/transformers/models/prophetnet/modeling_prophetnet.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from torch.nn import LayerNorm diff --git a/src/transformers/models/pvt/modeling_pvt.py b/src/transformers/models/pvt/modeling_pvt.py index 446a85944801..21af67542d70 100755 --- a/src/transformers/models/pvt/modeling_pvt.py +++ b/src/transformers/models/pvt/modeling_pvt.py @@ -23,7 +23,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/pvt_v2/modeling_pvt_v2.py b/src/transformers/models/pvt_v2/modeling_pvt_v2.py index e434223a94a2..204198787e45 100644 --- a/src/transformers/models/pvt_v2/modeling_pvt_v2.py +++ b/src/transformers/models/pvt_v2/modeling_pvt_v2.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/qwen2/modular_qwen2.py b/src/transformers/models/qwen2/modular_qwen2.py index b4555ef927e3..d946e4b7e1f0 100644 --- a/src/transformers/models/qwen2/modular_qwen2.py +++ b/src/transformers/models/qwen2/modular_qwen2.py @@ -1,7 +1,6 @@ from typing import Callable, Optional import torch -import torch.utils.checkpoint from packaging import version from torch import nn diff --git a/src/transformers/models/qwen2_5_omni/modular_qwen2_5_omni.py b/src/transformers/models/qwen2_5_omni/modular_qwen2_5_omni.py index 07cd851d4f88..afb0cda5ccfe 100644 --- a/src/transformers/models/qwen2_5_omni/modular_qwen2_5_omni.py +++ b/src/transformers/models/qwen2_5_omni/modular_qwen2_5_omni.py @@ -22,7 +22,6 @@ import numpy as np import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from torch.nn import Parameter diff --git a/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py b/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py index d62f94f37678..b59644c37df9 100644 --- a/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py +++ b/src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py @@ -25,7 +25,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from transformers.models.qwen2_vl.configuration_qwen2_vl import Qwen2VLConfig, Qwen2VLTextConfig from transformers.models.qwen2_vl.modeling_qwen2_vl import ( diff --git a/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py b/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py index 070eb6e89fd5..c4b37477d5af 100644 --- a/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py +++ b/src/transformers/models/qwen2_moe/modeling_qwen2_moe.py @@ -24,7 +24,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py b/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py index 0509ef9e085e..269f37492ad6 100644 --- a/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py +++ b/src/transformers/models/qwen2_vl/modeling_qwen2_vl.py @@ -25,7 +25,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from torch.nn import LayerNorm from ...activations import ACT2FN diff --git a/src/transformers/models/qwen3_moe/modular_qwen3_moe.py b/src/transformers/models/qwen3_moe/modular_qwen3_moe.py index b9213a5e5bbb..e7dd3dda00ac 100644 --- a/src/transformers/models/qwen3_moe/modular_qwen3_moe.py +++ b/src/transformers/models/qwen3_moe/modular_qwen3_moe.py @@ -18,7 +18,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/qwen3_next/modular_qwen3_next.py b/src/transformers/models/qwen3_next/modular_qwen3_next.py index f76a242877b9..e141e229eedf 100644 --- a/src/transformers/models/qwen3_next/modular_qwen3_next.py +++ b/src/transformers/models/qwen3_next/modular_qwen3_next.py @@ -18,7 +18,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/recurrent_gemma/modeling_recurrent_gemma.py b/src/transformers/models/recurrent_gemma/modeling_recurrent_gemma.py index daef714ab883..d7d1ce33e8f0 100644 --- a/src/transformers/models/recurrent_gemma/modeling_recurrent_gemma.py +++ b/src/transformers/models/recurrent_gemma/modeling_recurrent_gemma.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/regnet/modeling_regnet.py b/src/transformers/models/regnet/modeling_regnet.py index 5eb65d92b8be..70611113885f 100644 --- a/src/transformers/models/regnet/modeling_regnet.py +++ b/src/transformers/models/regnet/modeling_regnet.py @@ -18,7 +18,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/rembert/modeling_rembert.py b/src/transformers/models/rembert/modeling_rembert.py index a9ee455116c4..0fc9635cda88 100755 --- a/src/transformers/models/rembert/modeling_rembert.py +++ b/src/transformers/models/rembert/modeling_rembert.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/resnet/modeling_resnet.py b/src/transformers/models/resnet/modeling_resnet.py index c766a91cd277..59a509fe03cd 100644 --- a/src/transformers/models/resnet/modeling_resnet.py +++ b/src/transformers/models/resnet/modeling_resnet.py @@ -18,7 +18,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/roberta/modeling_roberta.py b/src/transformers/models/roberta/modeling_roberta.py index 6999dddf1b1a..33fb44118a90 100644 --- a/src/transformers/models/roberta/modeling_roberta.py +++ b/src/transformers/models/roberta/modeling_roberta.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/roberta_prelayernorm/modeling_roberta_prelayernorm.py b/src/transformers/models/roberta_prelayernorm/modeling_roberta_prelayernorm.py index 072466bd8b04..81481574b01e 100644 --- a/src/transformers/models/roberta_prelayernorm/modeling_roberta_prelayernorm.py +++ b/src/transformers/models/roberta_prelayernorm/modeling_roberta_prelayernorm.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/roc_bert/modeling_roc_bert.py b/src/transformers/models/roc_bert/modeling_roc_bert.py index e0b8b4b434ba..22a72f91bc38 100644 --- a/src/transformers/models/roc_bert/modeling_roc_bert.py +++ b/src/transformers/models/roc_bert/modeling_roc_bert.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/roformer/modeling_roformer.py b/src/transformers/models/roformer/modeling_roformer.py index 56dce4cb753b..3fc94cf87675 100644 --- a/src/transformers/models/roformer/modeling_roformer.py +++ b/src/transformers/models/roformer/modeling_roformer.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/rwkv/modeling_rwkv.py b/src/transformers/models/rwkv/modeling_rwkv.py index d86d4d0f8707..816b22f1b2dd 100644 --- a/src/transformers/models/rwkv/modeling_rwkv.py +++ b/src/transformers/models/rwkv/modeling_rwkv.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...generation import GenerationMixin diff --git a/src/transformers/models/sam2/modular_sam2.py b/src/transformers/models/sam2/modular_sam2.py index 5fff232a839c..be2a5eb1c6d2 100644 --- a/src/transformers/models/sam2/modular_sam2.py +++ b/src/transformers/models/sam2/modular_sam2.py @@ -21,7 +21,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from ...activations import ACT2FN from ...image_processing_utils import BatchFeature, get_size_dict diff --git a/src/transformers/models/sam2_video/modular_sam2_video.py b/src/transformers/models/sam2_video/modular_sam2_video.py index 83483e9d724e..9ba8e6526305 100644 --- a/src/transformers/models/sam2_video/modular_sam2_video.py +++ b/src/transformers/models/sam2_video/modular_sam2_video.py @@ -24,7 +24,6 @@ import torch import torch.nn as nn import torch.nn.functional as F -import torch.utils.checkpoint from torch import Tensor from tqdm import tqdm diff --git a/src/transformers/models/seamless_m4t/modeling_seamless_m4t.py b/src/transformers/models/seamless_m4t/modeling_seamless_m4t.py index 15f368281775..5078d437e978 100755 --- a/src/transformers/models/seamless_m4t/modeling_seamless_m4t.py +++ b/src/transformers/models/seamless_m4t/modeling_seamless_m4t.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/seamless_m4t_v2/modeling_seamless_m4t_v2.py b/src/transformers/models/seamless_m4t_v2/modeling_seamless_m4t_v2.py index ccad4450451d..7aa15cb84ddd 100644 --- a/src/transformers/models/seamless_m4t_v2/modeling_seamless_m4t_v2.py +++ b/src/transformers/models/seamless_m4t_v2/modeling_seamless_m4t_v2.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/segformer/modeling_segformer.py b/src/transformers/models/segformer/modeling_segformer.py index 4aa49d86466b..8a81f68beadd 100755 --- a/src/transformers/models/segformer/modeling_segformer.py +++ b/src/transformers/models/segformer/modeling_segformer.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss diff --git a/src/transformers/models/seggpt/modeling_seggpt.py b/src/transformers/models/seggpt/modeling_seggpt.py index 93ca95b91094..7e82d26c9e74 100644 --- a/src/transformers/models/seggpt/modeling_seggpt.py +++ b/src/transformers/models/seggpt/modeling_seggpt.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import functional as F diff --git a/src/transformers/models/sew/modular_sew.py b/src/transformers/models/sew/modular_sew.py index 5b4ee00c4a65..b15c2e5c23a3 100644 --- a/src/transformers/models/sew/modular_sew.py +++ b/src/transformers/models/sew/modular_sew.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/sew_d/modeling_sew_d.py b/src/transformers/models/sew_d/modeling_sew_d.py index 53eeda91f3da..f8b71241c79e 100644 --- a/src/transformers/models/sew_d/modeling_sew_d.py +++ b/src/transformers/models/sew_d/modeling_sew_d.py @@ -21,7 +21,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss, LayerNorm diff --git a/src/transformers/models/shieldgemma2/modeling_shieldgemma2.py b/src/transformers/models/shieldgemma2/modeling_shieldgemma2.py index 49261f039a56..5e12b0129ab6 100644 --- a/src/transformers/models/shieldgemma2/modeling_shieldgemma2.py +++ b/src/transformers/models/shieldgemma2/modeling_shieldgemma2.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from ...cache_utils import Cache from ...modeling_outputs import ImageClassifierOutputWithNoAttention diff --git a/src/transformers/models/smolvlm/modular_smolvlm.py b/src/transformers/models/smolvlm/modular_smolvlm.py index 25d55b1a974a..ffc7f06c97c9 100644 --- a/src/transformers/models/smolvlm/modular_smolvlm.py +++ b/src/transformers/models/smolvlm/modular_smolvlm.py @@ -16,7 +16,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...cache_utils import Cache, DynamicCache diff --git a/src/transformers/models/speecht5/modeling_speecht5.py b/src/transformers/models/speecht5/modeling_speecht5.py index 026585cdd77b..b3e79a46680c 100644 --- a/src/transformers/models/speecht5/modeling_speecht5.py +++ b/src/transformers/models/speecht5/modeling_speecht5.py @@ -19,7 +19,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, L1Loss diff --git a/src/transformers/models/splinter/modeling_splinter.py b/src/transformers/models/splinter/modeling_splinter.py index 905c7a27ad3d..116a17330923 100755 --- a/src/transformers/models/splinter/modeling_splinter.py +++ b/src/transformers/models/splinter/modeling_splinter.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/stablelm/modeling_stablelm.py b/src/transformers/models/stablelm/modeling_stablelm.py index 9daefe0a39df..6b31565a1b1d 100755 --- a/src/transformers/models/stablelm/modeling_stablelm.py +++ b/src/transformers/models/stablelm/modeling_stablelm.py @@ -23,7 +23,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/starcoder2/modular_starcoder2.py b/src/transformers/models/starcoder2/modular_starcoder2.py index a1baf7de8767..f37c75510abe 100644 --- a/src/transformers/models/starcoder2/modular_starcoder2.py +++ b/src/transformers/models/starcoder2/modular_starcoder2.py @@ -22,7 +22,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from transformers.utils.generic import check_model_inputs diff --git a/src/transformers/models/swiftformer/modeling_swiftformer.py b/src/transformers/models/swiftformer/modeling_swiftformer.py index 9e0c4c3147b7..95114e3d332c 100644 --- a/src/transformers/models/swiftformer/modeling_swiftformer.py +++ b/src/transformers/models/swiftformer/modeling_swiftformer.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2CLS diff --git a/src/transformers/models/swin/modeling_swin.py b/src/transformers/models/swin/modeling_swin.py index 37d3413fae5d..18b61abbd3a4 100644 --- a/src/transformers/models/swin/modeling_swin.py +++ b/src/transformers/models/swin/modeling_swin.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/swin2sr/modeling_swin2sr.py b/src/transformers/models/swin2sr/modeling_swin2sr.py index c2d12e8d78ae..e010a1d8a01e 100644 --- a/src/transformers/models/swin2sr/modeling_swin2sr.py +++ b/src/transformers/models/swin2sr/modeling_swin2sr.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/swinv2/modeling_swinv2.py b/src/transformers/models/swinv2/modeling_swinv2.py index 4d030178ed49..1463f0f82e7e 100644 --- a/src/transformers/models/swinv2/modeling_swinv2.py +++ b/src/transformers/models/swinv2/modeling_swinv2.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import Tensor, nn from ...activations import ACT2FN diff --git a/src/transformers/models/tapas/modeling_tapas.py b/src/transformers/models/tapas/modeling_tapas.py index 3f4bf53d1acc..075b834533b6 100644 --- a/src/transformers/models/tapas/modeling_tapas.py +++ b/src/transformers/models/tapas/modeling_tapas.py @@ -21,7 +21,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/timesformer/modeling_timesformer.py b/src/transformers/models/timesformer/modeling_timesformer.py index c0110b379aac..0aa06d5c33bb 100644 --- a/src/transformers/models/timesformer/modeling_timesformer.py +++ b/src/transformers/models/timesformer/modeling_timesformer.py @@ -19,7 +19,6 @@ import torch import torch.nn.functional -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/tvp/modeling_tvp.py b/src/transformers/models/tvp/modeling_tvp.py index 77d74bffe0ee..0b8b626d2dd2 100644 --- a/src/transformers/models/tvp/modeling_tvp.py +++ b/src/transformers/models/tvp/modeling_tvp.py @@ -19,7 +19,6 @@ from typing import Optional import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/univnet/modeling_univnet.py b/src/transformers/models/univnet/modeling_univnet.py index 1a9c2aa54e72..e7595ff38f8a 100644 --- a/src/transformers/models/univnet/modeling_univnet.py +++ b/src/transformers/models/univnet/modeling_univnet.py @@ -17,7 +17,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...modeling_outputs import ModelOutput diff --git a/src/transformers/models/video_llava/modeling_video_llava.py b/src/transformers/models/video_llava/modeling_video_llava.py index 41a4d0abed17..2db424455087 100644 --- a/src/transformers/models/video_llava/modeling_video_llava.py +++ b/src/transformers/models/video_llava/modeling_video_llava.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/videomae/modeling_videomae.py b/src/transformers/models/videomae/modeling_videomae.py index 335ae485289a..97c227f1d8bf 100755 --- a/src/transformers/models/videomae/modeling_videomae.py +++ b/src/transformers/models/videomae/modeling_videomae.py @@ -21,7 +21,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import MSELoss diff --git a/src/transformers/models/vilt/modeling_vilt.py b/src/transformers/models/vilt/modeling_vilt.py index ed749f85362a..75e58f9858fd 100755 --- a/src/transformers/models/vilt/modeling_vilt.py +++ b/src/transformers/models/vilt/modeling_vilt.py @@ -20,7 +20,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/visual_bert/modeling_visual_bert.py b/src/transformers/models/visual_bert/modeling_visual_bert.py index 3ea9b45537cb..cdc3e3adc69b 100755 --- a/src/transformers/models/visual_bert/modeling_visual_bert.py +++ b/src/transformers/models/visual_bert/modeling_visual_bert.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss, KLDivLoss, LogSoftmax diff --git a/src/transformers/models/vit/modeling_vit.py b/src/transformers/models/vit/modeling_vit.py index 3e84687d8ffa..d9c01927ffc4 100644 --- a/src/transformers/models/vit/modeling_vit.py +++ b/src/transformers/models/vit/modeling_vit.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/vit_mae/modeling_vit_mae.py b/src/transformers/models/vit_mae/modeling_vit_mae.py index 8b4b9efafeb5..a74d172805bd 100755 --- a/src/transformers/models/vit_mae/modeling_vit_mae.py +++ b/src/transformers/models/vit_mae/modeling_vit_mae.py @@ -21,7 +21,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/vit_msn/modeling_vit_msn.py b/src/transformers/models/vit_msn/modeling_vit_msn.py index fa99fa62a753..eee739b13864 100644 --- a/src/transformers/models/vit_msn/modeling_vit_msn.py +++ b/src/transformers/models/vit_msn/modeling_vit_msn.py @@ -18,7 +18,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/vitdet/modeling_vitdet.py b/src/transformers/models/vitdet/modeling_vitdet.py index 1498b72d856d..8debcaf11fa5 100644 --- a/src/transformers/models/vitdet/modeling_vitdet.py +++ b/src/transformers/models/vitdet/modeling_vitdet.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/vitpose/modeling_vitpose.py b/src/transformers/models/vitpose/modeling_vitpose.py index 900eb3aadf7b..250341797aab 100644 --- a/src/transformers/models/vitpose/modeling_vitpose.py +++ b/src/transformers/models/vitpose/modeling_vitpose.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...modeling_outputs import BackboneOutput diff --git a/src/transformers/models/vitpose_backbone/modeling_vitpose_backbone.py b/src/transformers/models/vitpose_backbone/modeling_vitpose_backbone.py index b5c596832fb4..1c61763d5e56 100644 --- a/src/transformers/models/vitpose_backbone/modeling_vitpose_backbone.py +++ b/src/transformers/models/vitpose_backbone/modeling_vitpose_backbone.py @@ -23,7 +23,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/vits/modeling_vits.py b/src/transformers/models/vits/modeling_vits.py index 45eca357ffb5..7300ea7f798e 100644 --- a/src/transformers/models/vits/modeling_vits.py +++ b/src/transformers/models/vits/modeling_vits.py @@ -20,7 +20,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/vivit/modeling_vivit.py b/src/transformers/models/vivit/modeling_vivit.py index aca26da16695..b27b56e640c6 100755 --- a/src/transformers/models/vivit/modeling_vivit.py +++ b/src/transformers/models/vivit/modeling_vivit.py @@ -17,7 +17,6 @@ from typing import Callable, Optional import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/wav2vec2/modeling_wav2vec2.py b/src/transformers/models/wav2vec2/modeling_wav2vec2.py index 90760d290e82..d8c58a333e07 100755 --- a/src/transformers/models/wav2vec2/modeling_wav2vec2.py +++ b/src/transformers/models/wav2vec2/modeling_wav2vec2.py @@ -21,7 +21,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/whisper/modeling_whisper.py b/src/transformers/models/whisper/modeling_whisper.py index da30a332d749..9ae3b33ebc6f 100644 --- a/src/transformers/models/whisper/modeling_whisper.py +++ b/src/transformers/models/whisper/modeling_whisper.py @@ -19,7 +19,6 @@ import numpy as np import torch -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/x_clip/modeling_x_clip.py b/src/transformers/models/x_clip/modeling_x_clip.py index 5a4b478ceef9..403b9a408162 100644 --- a/src/transformers/models/x_clip/modeling_x_clip.py +++ b/src/transformers/models/x_clip/modeling_x_clip.py @@ -19,7 +19,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/xglm/modeling_xglm.py b/src/transformers/models/xglm/modeling_xglm.py index cfa42502399b..0f863f3f274f 100755 --- a/src/transformers/models/xglm/modeling_xglm.py +++ b/src/transformers/models/xglm/modeling_xglm.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/xlm_roberta/modeling_xlm_roberta.py b/src/transformers/models/xlm_roberta/modeling_xlm_roberta.py index a398ff0b916c..a3a252572ec9 100644 --- a/src/transformers/models/xlm_roberta/modeling_xlm_roberta.py +++ b/src/transformers/models/xlm_roberta/modeling_xlm_roberta.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/xlm_roberta_xl/modeling_xlm_roberta_xl.py b/src/transformers/models/xlm_roberta_xl/modeling_xlm_roberta_xl.py index 99b925015a71..d0c71365d214 100644 --- a/src/transformers/models/xlm_roberta_xl/modeling_xlm_roberta_xl.py +++ b/src/transformers/models/xlm_roberta_xl/modeling_xlm_roberta_xl.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/xlstm/modeling_xlstm.py b/src/transformers/models/xlstm/modeling_xlstm.py index 7e2fce997683..5bb438efce7e 100644 --- a/src/transformers/models/xlstm/modeling_xlstm.py +++ b/src/transformers/models/xlstm/modeling_xlstm.py @@ -18,7 +18,6 @@ import torch import torch.nn.functional as F -import torch.utils.checkpoint from torch import nn from torch.nn import CrossEntropyLoss diff --git a/src/transformers/models/xmod/modeling_xmod.py b/src/transformers/models/xmod/modeling_xmod.py index bb1ba68d4624..7c8328447cb0 100644 --- a/src/transformers/models/xmod/modeling_xmod.py +++ b/src/transformers/models/xmod/modeling_xmod.py @@ -18,7 +18,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/yolos/modeling_yolos.py b/src/transformers/models/yolos/modeling_yolos.py index 2571cf82733d..13fd9886ea96 100755 --- a/src/transformers/models/yolos/modeling_yolos.py +++ b/src/transformers/models/yolos/modeling_yolos.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/yoso/modeling_yoso.py b/src/transformers/models/yoso/modeling_yoso.py index 221ebaa637fe..0ad53b81f492 100644 --- a/src/transformers/models/yoso/modeling_yoso.py +++ b/src/transformers/models/yoso/modeling_yoso.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/zamba/modeling_zamba.py b/src/transformers/models/zamba/modeling_zamba.py index 2f9edb1e113c..a69b7a0a3f86 100644 --- a/src/transformers/models/zamba/modeling_zamba.py +++ b/src/transformers/models/zamba/modeling_zamba.py @@ -23,7 +23,6 @@ from typing import Any, Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss diff --git a/src/transformers/models/zamba2/modular_zamba2.py b/src/transformers/models/zamba2/modular_zamba2.py index ea305ba7a5b6..3cada0c0dd43 100644 --- a/src/transformers/models/zamba2/modular_zamba2.py +++ b/src/transformers/models/zamba2/modular_zamba2.py @@ -19,7 +19,6 @@ from typing import Callable, Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN diff --git a/src/transformers/models/zoedepth/modeling_zoedepth.py b/src/transformers/models/zoedepth/modeling_zoedepth.py index d79bb27d56c0..7bbad31c2ee0 100644 --- a/src/transformers/models/zoedepth/modeling_zoedepth.py +++ b/src/transformers/models/zoedepth/modeling_zoedepth.py @@ -19,7 +19,6 @@ from typing import Optional, Union import torch -import torch.utils.checkpoint from torch import nn from ...activations import ACT2FN