Add tf code for efficientformer

huggingface · Apr 27, 2023 · 1d3c82d · 1d3c82d
1 parent 1670be4
commit 1d3c82d
Show file tree

Hide file tree

Showing 11 changed files with 1,579 additions and 4 deletions.
diff --git a/docs/source/en/index.mdx b/docs/source/en/index.mdx
@@ -303,7 +303,7 @@ Flax), PyTorch, and/or TensorFlow.
 |           DonutSwin           |       ❌       |       ❌       |       ✅        |         ❌         |      ❌      |
 |              DPR              |       ✅       |       ✅       |       ✅        |         ✅         |      ❌      |
 |              DPT              |       ❌       |       ❌       |       ✅        |         ❌         |      ❌      |
-|        EfficientFormer        |       ❌       |       ❌       |       ✅        |         ❌         |      ❌      |
+|        EfficientFormer        |       ❌       |       ❌       |       ✅        |         ✅         |      ❌      |
 |         EfficientNet          |       ❌       |       ❌       |       ✅        |         ❌         |      ❌      |
 |            ELECTRA            |       ✅       |       ✅       |       ✅        |         ✅         |      ✅      |
 |        Encoder decoder        |       ❌       |       ❌       |       ✅        |         ✅         |      ✅      |

diff --git a/docs/source/en/model_doc/efficientformer.mdx b/docs/source/en/model_doc/efficientformer.mdx
@@ -37,7 +37,7 @@ EfficientFormer-L7, obtains 83.3% accuracy with only 7.0 ms latency. Our work pr
 reach extremely low latency on mobile devices while maintaining high performance.*
 
 This model was contributed by [novice03](https://huggingface.co/novice03) and [Bearnardd](https://huggingface.co/Bearnardd).
-The original code can be found [here](https://github.com/snap-research/EfficientFormer).
+The original code can be found [here](https://github.com/snap-research/EfficientFormer). The TensorFlow version of this model was added by [D-Roberts](https://huggingface.co/D-Roberts).
 
 ## Documentation resources
 
@@ -66,3 +66,18 @@ The original code can be found [here](https://github.com/snap-research/Efficient
 
 [[autodoc]] EfficientFormerForImageClassificationWithTeacher
     - forward
+
+## TFEfficientFormerModel
+
+[[autodoc]] TFEfficientFormerModel
+    - call
+
+## TFEfficientFormerForImageClassification
+
+[[autodoc]] TFEfficientFormerForImageClassification
+    - call
+
+## TFEfficientFormerForImageClassificationWithTeacher
+
+[[autodoc]] TFEfficientFormerForImageClassificationWithTeacher
+    - call
diff --git a/src/transformers/__init__.py b/src/transformers/__init__.py
@@ -3041,6 +3041,15 @@
             "TFDPRReader",
         ]
     )
+    _import_structure["models.efficientformer"].extend(
+        [
+            "TF_EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+            "TFEfficientFormerForImageClassification",
+            "TFEfficientFormerForImageClassificationWithTeacher",
+            "TFEfficientFormerModel",
+            "TFEfficientFormerPreTrainedModel",
+        ]
+    )
     _import_structure["models.electra"].extend(
         [
             "TF_ELECTRA_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -6269,6 +6278,13 @@
             TFDPRQuestionEncoder,
             TFDPRReader,
         )
+        from .models.efficientformer import (
+            TF_EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+            TFEfficientFormerForImageClassification,
+            TFEfficientFormerForImageClassificationWithTeacher,
+            TFEfficientFormerModel,
+            TFEfficientFormerPreTrainedModel,
+        )
         from .models.electra import (
             TF_ELECTRA_PRETRAINED_MODEL_ARCHIVE_LIST,
             TFElectraForMaskedLM,

diff --git a/src/transformers/models/auto/modeling_tf_auto.py b/src/transformers/models/auto/modeling_tf_auto.py
@@ -47,6 +47,7 @@
         ("deit", "TFDeiTModel"),
         ("distilbert", "TFDistilBertModel"),
         ("dpr", "TFDPRQuestionEncoder"),
+        ("efficientformer", "TFEfficientFormerModel"),
         ("electra", "TFElectraModel"),
         ("esm", "TFEsmModel"),
         ("flaubert", "TFFlaubertModel"),
@@ -201,6 +202,10 @@
         ("cvt", "TFCvtForImageClassification"),
         ("data2vec-vision", "TFData2VecVisionForImageClassification"),
         ("deit", ("TFDeiTForImageClassification", "TFDeiTForImageClassificationWithTeacher")),
+        (
+            "efficientformer",
+            ("TFEfficientFormerForImageClassification", "TFEfficientFormerForImageClassificationWithTeacher"),
+        ),
         ("mobilevit", "TFMobileViTForImageClassification"),
         ("regnet", "TFRegNetForImageClassification"),
         ("resnet", "TFResNetForImageClassification"),

diff --git a/src/transformers/models/efficientformer/__init__.py b/src/transformers/models/efficientformer/__init__.py
@@ -13,7 +13,13 @@
 # limitations under the License.
 from typing import TYPE_CHECKING
 
-from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available, is_vision_available
+from ...utils import (
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    is_tf_available,
+    is_torch_available,
+    is_vision_available,
+)
 
 
 _import_structure = {
@@ -45,6 +51,20 @@
         "EfficientFormerPreTrainedModel",
     ]
 
+try:
+    if not is_tf_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_tf_efficientformer"] = [
+        "TF_EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST",
+        "TFEfficientFormerForImageClassification",
+        "TFEfficientFormerForImageClassificationWithTeacher",
+        "TFEfficientFormerModel",
+        "TFEfficientFormerPreTrainedModel",
+    ]
+
 if TYPE_CHECKING:
     from .configuration_efficientformer import EFFICIENTFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, EfficientFormerConfig
 
@@ -69,6 +89,19 @@
             EfficientFormerModel,
             EfficientFormerPreTrainedModel,
         )
+    try:
+        if not is_tf_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_tf_efficientformer import (
+            TF_EFFICIENTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
+            TFEfficientFormerForImageClassification,
+            TFEfficientFormerForImageClassificationWithTeacher,
+            TFEfficientFormerModel,
+            TFEfficientFormerPreTrainedModel,
+        )
 
 else:
     import sys

diff --git a/src/transformers/models/efficientformer/configuration_efficientformer.py b/src/transformers/models/efficientformer/configuration_efficientformer.py
@@ -91,6 +91,8 @@ class EfficientFormerConfig(PretrainedConfig):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         layer_norm_eps (`float`, *optional*, defaults to 1e-12):
             The epsilon used by the layer normalization layers.
+        image_size (`int`, *optional*, defaults to `224`):
+            The size (resolution) of each image.
 
     Example:
 
@@ -136,6 +138,7 @@ def __init__(
         hidden_act: str = "gelu",
         initializer_range: float = 0.02,
         layer_norm_eps: float = 1e-12,
+        image_size: int = 224,
         **kwargs,
     ) -> None:
         super().__init__(**kwargs)
@@ -165,3 +168,4 @@ def __init__(
         self.distillation = distillation
         self.use_layer_scale = use_layer_scale
         self.layer_scale_init_value = layer_scale_init_value
+        self.image_size = image_size