Refactor to allow for a wider model in TIMM (#3976)

* update for releases 2.2.0rc0 * Fix Classification explain forward issue (#3867) Fix bug * Fix e2e code error (#3871) * Update test_cli.py * Update tests/e2e/cli/test_cli.py Co-authored-by: Eunwoo Shin <eunwoo.shin@intel.com> * Update test_cli.py * Update test_cli.py --------- Co-authored-by: Eunwoo Shin <eunwoo.shin@intel.com> * Add documentation about configurable input size (#3870) * add docs about configurable input size * update api usecase and fix bug * Fix zero-shot e2e (#3876) Fix * Fix DeiT for multi-label classification (#3881) Remove init_args * Fix Semi-SL for ViT accuracy drop (#3883) Remove init_args * Update docs for 2.2 (#3884) Update docs * Fix mean and scale for segmentation task (#3885) fix mean and scale * Update MAPI in 2.2 (#3889) * Bump MAPI * Update exportable code requirements * Improve Semi-SL for LiteHRNet (small-medium case) (#3891) * change drop pixels value * go safe, change only tested models * minor * Improve h-cls for eff models (#3893) * Update step size for eff v2 * Update effb0 recipe * Fix maskrcnn swin nncf acc drop (#3900) update maskrcnn swimt model type to transformer * Add keypoint detection recipe for single object cases (#3903) * add rtmpose_tiny for single obj * add rtmpose_tiny for single obj * modify test subset name * fix unit test * update recipe with reset * Improve acc drop of efficientnetv2 for h-label cls (#3907) * Add warmup_iters for effv2 * Update max_epochs * Fix pretrained weight cached dir for timm (#3909) * Fix pretrained_weight for timm * Fix unit-test * Fix keypoint detection single obj recipe (#3915) * add rtmpose_tiny for single obj * modify test subset name * fix unit test * property for pck * Fix cached dir for timm & hugging-face (#3914) * Fix cached dir * Pretrained weight download unit-test * Fix pre-commit * Fix wrong template id mapping for anomaly (#3916) * Update script to allow setting otx version using env. variable (#3913) * Fix Datamodule creation for OV in AutoConfigurator (#3920) Fix datamodule for ov * Update tpp file for 2.2.0 (#3921) * Fix names for ignored scope [HOT-FIX, 2.2.0] (#3924) fix names for ignored scope * Fix classification rt_info (#3922) * Restore output_raw_scores for classificaiton * Add uts * Fix linter * Update label info (#3925) add label info to init Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> * Fix binary classification metric task (#3928) * Fix binary classification * Add unit-tests * Improve MaskRCNN SwinT NNCF (#3929) * ignore heads and disable smooth quant * add activations_range_estimator_params * update changelog * Fix get_item for Chained Tasks in Classification (#3931) * Fix Task Chain * Add multi-label case as well * Add multi-label case as well2 * Add H-label case * Correct Keyerror for h-label cls in label_groups for dm_label_categories using label's id/key (#3932) Modify label_groups for dm_label_categories with id/key of label * Remove datumaro attribute id from tiling, add subset names (#3933) * remove datumaro attribute id from tiling * add subset names * Fix soft predictions for Semantic Segmentation (#3934) fix soft preds * Update STFPM config (#3935) * Add missing pretrained weights when creating a docker image (#3938) * Fix pre-trained weight downloader * Remove if condition for pretrained wiehgt download * Change default option 'full' to 'base' in otx install (#3937) * Change option full to base for otx install * Fix wrong code * Fix issue * Fix docs * Fix auto adapt batch size in Converter (#3939) * Enable auto adapt batch size into converter * Fix wrong * Fix hpo converter (#3940) * save best hp after hpo * add test * Fix tiling XAI out of range (#3943) - Fix tile merge XAI out of range * enable model export (#3952) Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> * Move templates from OTX1.X to OTX2.X (#3951) * add otx1.6 templates * added new models * delete entrypoints and nncf cfg * updated some hyperparams * fix for rtmdet_tiny * updated converter * Update classification templates * Update det, r-det, vpm * Update template.yaml * changed warmaup value in train.yaml --------- Co-authored-by: Kang, Harim <harim.kang@intel.com> Co-authored-by: Kim, Sungchul <sungchul.kim@intel.com> * Add missing tile recipes and various tile recipe changes (#3942) * add missing tile recipes * Fix tiling XAI out of range (#3943) - Fix tile merge XAI out of range * update xai tile merge * update rtdetr * update tile recipes * update rtdetr tile postprocess * update rtdetr recipes and tile recipes * update tile recipes * fix rtdetr unittest * update recipes * refactor tile unit test * address pr reviews * remove unnecessary files * update color channel * fix image channel passing * include tiling in cli integration test * remove transform_bbox --------- Co-authored-by: Vladislav Sovrasov <sovrasov.vlad@gmail.com> * Support ImageFromBytes (#3948) * add image_from_bytes Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> * refactor code Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> * allow empty anomalous masks Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> --------- Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> * Change categories mapping logic (#3946) * change pre-filtering logic * Update src/otx/core/data/pre_filtering.py Co-authored-by: Eunwoo Shin <eunwoo.shin@intel.com> --------- Co-authored-by: Eunwoo Shin <eunwoo.shin@intel.com> * Update for 2.2.0rc1 (#3956) * Include Geti arrow dataset subset names (#3962) * restrited number of output masks by tiling * add geti subset name * update num of max pred * Include full image with anno in case there's no tile in tile dataset (#3964) * include full image with anno incase there's no tile in dataset * update test * Add type checker in converter for callable functions (optimizer, scheduler) (#3968) Fix converter callable functions (optimizer, scheduler) * Update for 2.2.0rc2 (#3969) update for 2.2.0rc2 * Refactor TIMM * Remove experimental recipes * Revert timm version * Fix conflict2 * Fix unit-test --------- Signed-off-by: Ashwin Vaidya <ashwinnitinvaidya@gmail.com> Co-authored-by: Yunchu Lee <yunchu.lee@intel.com> Co-authored-by: Emily Chun <emily.chun@intel.com> Co-authored-by: Eunwoo Shin <eunwoo.shin@intel.com> Co-authored-by: Kim, Sungchul <sungchul.kim@intel.com> Co-authored-by: Prokofiev Kirill <kirill.prokofiev@intel.com> Co-authored-by: Vladislav Sovrasov <sovrasov.vlad@gmail.com> Co-authored-by: Sooah Lee <sooah.lee@intel.com> Co-authored-by: Eugene Liu <eugene.liu@intel.com> Co-authored-by: Wonju Lee <wonju.lee@intel.com> Co-authored-by: Ashwin Vaidya <ashwin.vaidya@intel.com>
openvinotoolkit · Sep 27, 2024 · 00d5604 · 00d5604
1 parent a25a94f
commit 00d5604
Show file tree

Hide file tree

Showing 8 changed files with 112 additions and 51 deletions.
diff --git a/src/otx/algo/classification/backbones/timm.py b/src/otx/algo/classification/backbones/timm.py
@@ -1,57 +1,47 @@
 # Copyright (C) 2024 Intel Corporation
 # SPDX-License-Identifier: Apache-2.0
 
-"""EfficientNetV2 model.
+"""Timm Backbone Class for OTX classification.
 
 Original papers:
 - 'EfficientNetV2: Smaller Models and Faster Training,' https://arxiv.org/abs/2104.00298,
 - 'Adversarial Examples Improve Image Recognition,' https://arxiv.org/abs/1911.09665.
 """
 from __future__ import annotations
 
-from typing import Literal
-
 import timm
 import torch
 from torch import nn
 
-TimmModelType = Literal[
-    "mobilenetv3_large_100_miil_in21k",
-    "mobilenetv3_large_100_miil",
-    "tresnet_m",
-    "tf_efficientnetv2_s.in21k",
-    "tf_efficientnetv2_s.in21ft1k",
-    "tf_efficientnetv2_m.in21k",
-    "tf_efficientnetv2_m.in21ft1k",
-    "tf_efficientnetv2_b0",
-]
-
 
 class TimmBackbone(nn.Module):
-    """Timm backbone model."""
+    """Timm backbone model.
+
+    Args:
+        model_name (str): The name of the model.
+            You can find available models at timm.list_models() or timm.list_pretrained().
+        pretrained (bool, optional): Whether to load pretrained weights. Defaults to False.
+    """
 
     def __init__(
         self,
-        backbone: TimmModelType,
+        model_name: str,
         pretrained: bool = False,
-        pooling_type: str = "avg",
         **kwargs,
     ):
         super().__init__(**kwargs)
-        self.backbone = backbone
+        self.model_name = model_name
         self.pretrained: bool | dict = pretrained
-        self.is_mobilenet = backbone.startswith("mobilenet")
 
         self.model = timm.create_model(
-            self.backbone,
+            self.model_name,
             pretrained=pretrained,
             num_classes=1000,
         )
 
         self.model.classifier = None  # Detach classifier. Only use 'backbone' part in otx.
         self.num_head_features = self.model.num_features
-        self.num_features = self.model.conv_head.in_channels if self.is_mobilenet else self.model.num_features
-        self.pooling_type = pooling_type
+        self.num_features = self.model.num_features
 
     def forward(self, x: torch.Tensor, **kwargs) -> tuple[torch.Tensor]:
         """Forward."""
@@ -60,11 +50,6 @@ def forward(self, x: torch.Tensor, **kwargs) -> tuple[torch.Tensor]:
 
     def extract_features(self, x: torch.Tensor) -> torch.Tensor:
         """Extract features."""
-        if self.is_mobilenet:
-            x = self.model.conv_stem(x)
-            x = self.model.bn1(x)
-            x = self.model.act1(x)
-            return self.model.blocks(x)
         return self.model.forward_features(x)
 
     def get_config_optim(self, lrs: list[float] | float) -> list[dict[str, float]]:

diff --git a/src/otx/algo/classification/timm_model.py b/src/otx/algo/classification/timm_model.py
@@ -1,7 +1,7 @@
 # Copyright (C) 2024 Intel Corporation
 # SPDX-License-Identifier: Apache-2.0
 
-"""EfficientNetV2 model implementation."""
+"""TIMM wrapper model class for OTX."""
 
 from __future__ import annotations
 
@@ -12,7 +12,7 @@
 import torch
 from torch import nn
 
-from otx.algo.classification.backbones.timm import TimmBackbone, TimmModelType
+from otx.algo.classification.backbones.timm import TimmBackbone
 from otx.algo.classification.classifier import HLabelClassifier, ImageClassifier, SemiSLClassifier
 from otx.algo.classification.heads import (
     HierarchicalCBAMClsHead,
@@ -50,12 +50,38 @@
 
 
 class TimmModelForMulticlassCls(OTXMulticlassClsModel):
-    """TimmModel for multi-class classification task."""
+    """TimmModel for multi-class classification task.
+
+    Args:
+        label_info (LabelInfoTypes): The label information for the classification task.
+        model_name (str): The name of the model.
+            You can find available models at timm.list_models() or timm.list_pretrained().
+        input_size (tuple[int, int], optional): Model input size in the order of height and width.
+            Defaults to (224, 224).
+        pretrained (bool, optional): Whether to load pretrained weights. Defaults to True.
+        optimizer (OptimizerCallable, optional): The optimizer callable for training the model.
+        scheduler (LRSchedulerCallable | LRSchedulerListCallable, optional): The learning rate scheduler callable.
+        metric (MetricCallable, optional): The metric callable for evaluating the model.
+            Defaults to MultiClassClsMetricCallable.
+        torch_compile (bool, optional): Whether to compile the model using TorchScript. Defaults to False.
+        train_type (Literal[OTXTrainType.SUPERVISED, OTXTrainType.SEMI_SUPERVISED], optional): The training type.
+
+    Example:
+        1. API
+            >>> model = TimmModelForMulticlassCls(
+            ...     model_name="tf_efficientnetv2_s.in21k",
+            ...     label_info=<Number-of-classes>,
+            ... )
+        2. CLI
+            >>> otx train \
+            ... --model otx.algo.classification.timm_model.TimmModelForMulticlassCls \
+            ... --model.model_name tf_efficientnetv2_s.in21k
+    """
 
     def __init__(
         self,
         label_info: LabelInfoTypes,
-        backbone: TimmModelType,
+        model_name: str,
         input_size: tuple[int, int] = (224, 224),  # input size of default classification data recipe
         pretrained: bool = True,
         optimizer: OptimizerCallable = DefaultOptimizerCallable,
@@ -64,7 +90,7 @@ def __init__(
         torch_compile: bool = False,
         train_type: Literal[OTXTrainType.SUPERVISED, OTXTrainType.SEMI_SUPERVISED] = OTXTrainType.SUPERVISED,
     ) -> None:
-        self.backbone = backbone
+        self.model_name = model_name
         self.pretrained = pretrained
 
         super().__init__(
@@ -92,7 +118,7 @@ def _create_model(self) -> nn.Module:
         return model
 
     def _build_model(self, num_classes: int) -> nn.Module:
-        backbone = TimmBackbone(backbone=self.backbone, pretrained=self.pretrained)
+        backbone = TimmBackbone(model_name=self.model_name, pretrained=self.pretrained)
         neck = GlobalAveragePooling(dim=2)
         if self.train_type == OTXTrainType.SEMI_SUPERVISED:
             return SemiSLClassifier(
@@ -142,20 +168,45 @@ def forward_for_tracing(self, image: torch.Tensor) -> torch.Tensor | dict[str, t
 
 
 class TimmModelForMultilabelCls(OTXMultilabelClsModel):
-    """TimmModel for multi-label classification task."""
+    """TimmModel for multi-label classification task.
+
+    Args:
+        label_info (LabelInfoTypes): The label information for the classification task.
+        model_name (str): The name of the model.
+            You can find available models at timm.list_models() or timm.list_pretrained().
+        input_size (tuple[int, int], optional): Model input size in the order of height and width.
+            Defaults to (224, 224).
+        pretrained (bool, optional): Whether to load pretrained weights. Defaults to True.
+        optimizer (OptimizerCallable, optional): The optimizer callable for training the model.
+        scheduler (LRSchedulerCallable | LRSchedulerListCallable, optional): The learning rate scheduler callable.
+        metric (MetricCallable, optional): The metric callable for evaluating the model.
+            Defaults to MultiLabelClsMetricCallable.
+        torch_compile (bool, optional): Whether to compile the model using TorchScript. Defaults to False.
+
+    Example:
+        1. API
+            >>> model = TimmModelForMultilabelCls(
+            ...     model_name="tf_efficientnetv2_s.in21k",
+            ...     label_info=<Number-of-classes>,
+            ... )
+        2. CLI
+            >>> otx train \
+            ... --model otx.algo.classification.timm_model.TimmModelForMultilabelCls \
+            ... --model.model_name tf_efficientnetv2_s.in21k
+    """
 
     def __init__(
         self,
         label_info: LabelInfoTypes,
-        backbone: TimmModelType,
+        model_name: str,
         input_size: tuple[int, int] = (224, 224),  # input size of default classification data recipe
         pretrained: bool = True,
         optimizer: OptimizerCallable = DefaultOptimizerCallable,
         scheduler: LRSchedulerCallable | LRSchedulerListCallable = DefaultSchedulerCallable,
         metric: MetricCallable = MultiLabelClsMetricCallable,
         torch_compile: bool = False,
     ) -> None:
-        self.backbone = backbone
+        self.model_name = model_name
         self.pretrained = pretrained
 
         super().__init__(
@@ -182,7 +233,7 @@ def _create_model(self) -> nn.Module:
         return model
 
     def _build_model(self, num_classes: int) -> nn.Module:
-        backbone = TimmBackbone(backbone=self.backbone, pretrained=self.pretrained)
+        backbone = TimmBackbone(model_name=self.model_name, pretrained=self.pretrained)
         return ImageClassifier(
             backbone=backbone,
             neck=GlobalAveragePooling(dim=2),
@@ -222,22 +273,47 @@ def forward_for_tracing(self, image: torch.Tensor) -> torch.Tensor | dict[str, t
 
 
 class TimmModelForHLabelCls(OTXHlabelClsModel):
-    """EfficientNetV2 Model for hierarchical label classification task."""
+    """Timm Model for hierarchical label classification task.
+
+    Args:
+        label_info (HLabelInfo): The label information for the classification task.
+        model_name (str): The name of the model.
+            You can find available models at timm.list_models() or timm.list_pretrained().
+        input_size (tuple[int, int], optional): Model input size in the order of height and width.
+            Defaults to (224, 224).
+        pretrained (bool, optional): Whether to load pretrained weights. Defaults to True.
+        optimizer (OptimizerCallable, optional): The optimizer callable for training the model.
+        scheduler (LRSchedulerCallable | LRSchedulerListCallable, optional): The learning rate scheduler callable.
+        metric (MetricCallable, optional): The metric callable for evaluating the model.
+            Defaults to HLabelClsMetricCallable.
+        torch_compile (bool, optional): Whether to compile the model using TorchScript. Defaults to False.
+
+    Example:
+        1. API
+            >>> model = TimmModelForHLabelCls(
+            ...     model_name="tf_efficientnetv2_s.in21k",
+            ...     label_info=<h-label-info>,
+            ... )
+        2. CLI
+            >>> otx train \
+            ... --model otx.algo.classification.timm_model.TimmModelForHLabelCls \
+            ... --model.model_name tf_efficientnetv2_s.in21k
+    """
 
     label_info: HLabelInfo
 
     def __init__(
         self,
         label_info: HLabelInfo,
-        backbone: TimmModelType,
+        model_name: str,
         input_size: tuple[int, int] = (224, 224),  # input size of default classification data recipe
         pretrained: bool = True,
         optimizer: OptimizerCallable = DefaultOptimizerCallable,
         scheduler: LRSchedulerCallable | LRSchedulerListCallable = DefaultSchedulerCallable,
         metric: MetricCallable = HLabelClsMetricCallable,
         torch_compile: bool = False,
     ) -> None:
-        self.backbone = backbone
+        self.model_name = model_name
         self.pretrained = pretrained
 
         super().__init__(
@@ -267,7 +343,7 @@ def _create_model(self) -> nn.Module:
         return model
 
     def _build_model(self, head_config: dict) -> nn.Module:
-        backbone = TimmBackbone(backbone=self.backbone, pretrained=self.pretrained)
+        backbone = TimmBackbone(model_name=self.model_name, pretrained=self.pretrained)
         copied_head_config = copy(head_config)
         copied_head_config["step_size"] = (ceil(self.input_size[0] / 32), ceil(self.input_size[1] / 32))
         return HLabelClassifier(

diff --git a/src/otx/recipe/classification/h_label_cls/efficientnet_v2.yaml b/src/otx/recipe/classification/h_label_cls/efficientnet_v2.yaml
@@ -1,7 +1,7 @@
 model:
   class_path: otx.algo.classification.timm_model.TimmModelForHLabelCls
   init_args:
-    backbone: tf_efficientnetv2_s.in21k
+    model_name: tf_efficientnetv2_s.in21k
 
     optimizer:
       class_path: torch.optim.SGD

diff --git a/src/otx/recipe/classification/multi_class_cls/efficientnet_v2.yaml b/src/otx/recipe/classification/multi_class_cls/efficientnet_v2.yaml
@@ -2,7 +2,7 @@ model:
   class_path: otx.algo.classification.timm_model.TimmModelForMulticlassCls
   init_args:
     label_info: 1000
-    backbone: tf_efficientnetv2_s.in21k
+    model_name: tf_efficientnetv2_s.in21k
 
     optimizer:
       class_path: torch.optim.SGD

diff --git a/src/otx/recipe/classification/multi_class_cls/semisl/efficientnet_v2_semisl.yaml b/src/otx/recipe/classification/multi_class_cls/semisl/efficientnet_v2_semisl.yaml
@@ -2,7 +2,7 @@ model:
   class_path: otx.algo.classification.timm_model.TimmModelForMulticlassCls
   init_args:
     label_info: 1000
-    backbone: tf_efficientnetv2_s.in21k
+    model_name: tf_efficientnetv2_s.in21k
     train_type: SEMI_SUPERVISED
 
     optimizer:

diff --git a/src/otx/recipe/classification/multi_label_cls/efficientnet_v2.yaml b/src/otx/recipe/classification/multi_label_cls/efficientnet_v2.yaml
@@ -2,7 +2,7 @@ model:
   class_path: otx.algo.classification.timm_model.TimmModelForMultilabelCls
   init_args:
     label_info: 1000
-    backbone: tf_efficientnetv2_s.in21k
+    model_name: tf_efficientnetv2_s.in21k
 
     optimizer:
       class_path: torch.optim.SGD

diff --git a/tests/unit/algo/classification/backbones/test_timm.py b/tests/unit/algo/classification/backbones/test_timm.py
@@ -11,11 +11,11 @@
 
 class TestOTXEfficientNetV2:
     def test_forward(self):
-        model = TimmBackbone(backbone="tf_efficientnetv2_s.in21k")
+        model = TimmBackbone(model_name="tf_efficientnetv2_s.in21k")
         assert model(torch.randn(1, 3, 244, 244))[0].shape == torch.Size([1, 1280, 8, 8])
 
     def test_get_config_optim(self):
-        model = TimmBackbone(backbone="tf_efficientnetv2_s.in21k")
+        model = TimmBackbone(model_name="tf_efficientnetv2_s.in21k")
         assert model.get_config_optim([0.01])[0]["lr"] == 0.01
         assert model.get_config_optim(0.01)[0]["lr"] == 0.01
 
@@ -24,5 +24,5 @@ def test_check_pretrained_weight_download(self):
         if target.exists():
             shutil.rmtree(target)
         assert not target.exists()
-        TimmBackbone(backbone="tf_efficientnetv2_s.in21k", pretrained=True)
+        TimmBackbone(model_name="tf_efficientnetv2_s.in21k", pretrained=True)
         assert target.exists()
diff --git a/tests/unit/algo/classification/test_timm_model.py b/tests/unit/algo/classification/test_timm_model.py
@@ -21,7 +21,7 @@
 def fxt_multi_class_cls_model():
     return TimmModelForMulticlassCls(
         label_info=10,
-        backbone="tf_efficientnetv2_s.in21k",
+        model_name="tf_efficientnetv2_s.in21k",
     )
 
 
@@ -59,7 +59,7 @@ def test_predict_step(self, fxt_multi_class_cls_model, fxt_multiclass_cls_batch_
 def fxt_multi_label_cls_model():
     return TimmModelForMultilabelCls(
         label_info=10,
-        backbone="tf_efficientnetv2_s.in21k",
+        model_name="tf_efficientnetv2_s.in21k",
     )
 
 
@@ -97,7 +97,7 @@ def test_predict_step(self, fxt_multi_label_cls_model, fxt_multilabel_cls_batch_
 def fxt_h_label_cls_model(fxt_hlabel_cifar):
     return TimmModelForHLabelCls(
         label_info=fxt_hlabel_cifar,
-        backbone="tf_efficientnetv2_s.in21k",
+        model_name="tf_efficientnetv2_s.in21k",
     )