Upload folder using huggingface_hub

modeling_MMRet_CLIP.py

@@ -22,12 +22,12 @@ import torch.utils.checkpoint
 from torch import nn
 from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 from PIL import Image
-from ...activations import ACT2FN
-from ...modeling_attn_mask_utils import _create_4d_causal_attention_mask, _prepare_4d_attention_mask
-from ...modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling, ImageClassifierOutput
-from ...modeling_utils import PreTrainedModel
-from ...pytorch_utils import is_torch_greater_or_equal_than_2_2
-from ...utils import (
+from transformers.activations import ACT2FN
+from transformers.modeling_attn_mask_utils import _create_4d_causal_attention_mask, _prepare_4d_attention_mask
+from transformers.modeling_outputs import BaseModelOutput, BaseModelOutputWithPooling, ImageClassifierOutput
+from transformers.modeling_utils import PreTrainedModel
+from transformers.pytorch_utils import is_torch_greater_or_equal_than_2_2
+from transformers.utils import (
     ModelOutput,
     add_code_sample_docstrings,
     add_start_docstrings,
@@ -37,7 +37,7 @@ from ...utils import (
     logging,
     replace_return_docstrings,
 )
-from .configuration_clip import CLIPConfig, CLIPTextConfig, CLIPVisionConfig
+from transformers.models.clip.configuration_clip import CLIPConfig, CLIPTextConfig, CLIPVisionConfig
 
 
 if is_flash_attn_2_available():
@@ -47,11 +47,10 @@ if is_flash_attn_2_available():
 logger = logging.get_logger(__name__)
 
 # General docstring
-_CONFIG_FOR_DOC = "CLIPConfig"
-_CHECKPOINT_FOR_DOC = "openai/clip-vit-base-patch32"
+_CONFIG_FOR_DOC = "MMRet_CLIP"
 
 # Image classification docstring
-_IMAGE_CLASS_CHECKPOINT = "openai/clip-vit-base-patch32"
+_IMAGE_CLASS_CHECKPOINT = "JUNJIE/MMRet-large"
 _IMAGE_CLASS_EXPECTED_OUTPUT = "LABEL_0"