Docs: avoid duplicate target links (#1299)

* Docs: avoid duplicate target links

Signed-off-by: Adam J. Stewart <ajstewart426@gmail.com>

* Correct format of anonymous hyperlink target

Signed-off-by: Adam J. Stewart <ajstewart426@gmail.com>

* Update all remaining links

Signed-off-by: Adam J. Stewart <ajstewart426@gmail.com>

* Display colab badge links

Signed-off-by: Adam J. Stewart <ajstewart426@gmail.com>

* Document more decoders

Signed-off-by: Adam J. Stewart <ajstewart426@gmail.com>

---------

Signed-off-by: Adam J. Stewart <ajstewart426@gmail.com>

Author: adamjstewart

docs/quickstart.rst

@@ -6,7 +6,7 @@
 Segmentation model is just a PyTorch nn.Module, which can be created as easy as:
 
 .. code-block:: python
-    
+
     import segmentation_models_pytorch as smp
 
     model = smp.Unet(
@@ -65,5 +65,5 @@ Check the following examples:
    :target: https://colab.research.google.com/github/qubvel/segmentation_models.pytorch/blob/main/examples/binary_segmentation_intro.ipynb
    :alt: Open In Colab
 
-- Finetuning notebook on Oxford Pet dataset with `PyTorch Lightning <https://github.com/qubvel/segmentation_models.pytorch/blob/main/examples/binary_segmentation_intro.ipynb>`_ |colab-badge|
-- Finetuning script for cloth segmentation with `PyTorch Lightning <https://github.com/ternaus/cloths_segmentation>`_
+- Finetuning notebook on Oxford Pet dataset with `PyTorch Lightning <https://github.com/qubvel/segmentation_models.pytorch/blob/main/examples/binary_segmentation_intro.ipynb>`__ |colab-badge|
+- Finetuning script for cloth segmentation with `PyTorch Lightning <https://github.com/ternaus/cloths_segmentation>`__

docs/save_load.rst

@@ -70,8 +70,8 @@ For example:
 Saving with preprocessing transform (Albumentations)
 ----------------------------------------------------
 
-You can save the preprocessing transform along with the model and push it to the Hub. 
-This can be useful when you want to share the model with the preprocessing transform that was used during training, 
+You can save the preprocessing transform along with the model and push it to the Hub.
+This can be useful when you want to share the model with the preprocessing transform that was used during training,
 to make sure that the inference pipeline is consistent with the training pipeline.
 
 .. code:: python
@@ -104,12 +104,13 @@ Conclusion
 
 By following these steps, you can easily save, share, and load your models, facilitating collaboration and reproducibility in your projects. Don't forget to replace the placeholders with your actual model paths and names.
 
-|colab-badge|
+|binary-segmentation-intro|
+|save-load-model-and-share-with-hf-hub|
 
-.. |colab-badge| image:: https://colab.research.google.com/assets/colab-badge.svg
+.. |binary-segmentation-intro| image:: https://colab.research.google.com/assets/colab-badge.svg
     :target: https://colab.research.google.com/github/qubvel/segmentation_models.pytorch/blob/main/examples/binary_segmentation_intro.ipynb
     :alt: Open In Colab
 
-.. |colab-badge| image:: https://colab.research.google.com/assets/colab-badge.svg
+.. |save-load-model-and-share-with-hf-hub| image:: https://colab.research.google.com/assets/colab-badge.svg
     :target: https://colab.research.google.com/github/qubvel/segmentation_models.pytorch/blob/main/examples/save_load_model_and_share_with_hf_hub.ipynb
     :alt: Open In Colab

segmentation_models_pytorch/decoders/deeplabv3/model.py

@@ -14,7 +14,7 @@
 
 
 class DeepLabV3(SegmentationModel):
-    """DeepLabV3_ implementation from "Rethinking Atrous Convolution for Semantic Image Segmentation"
+    """`DeepLabV3`__ implementation from "Rethinking Atrous Convolution for Semantic Image Segmentation"
 
     Args:
         encoder_name: Name of the classification model that will be used as an encoder (a.k.a backbone)
@@ -52,9 +52,7 @@ class DeepLabV3(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **DeepLabV3**
 
-    .. _DeeplabV3:
-        https://arxiv.org/abs/1706.05587
-
+    .. __: https://arxiv.org/abs/1706.05587
     """
 
     @supports_config_loading
@@ -141,7 +139,7 @@ def load_state_dict(self, state_dict, *args, **kwargs):
 
 
 class DeepLabV3Plus(SegmentationModel):
-    """DeepLabV3+ implementation from "Encoder-Decoder with Atrous Separable
+    """`DeepLabV3+`__ implementation from "Encoder-Decoder with Atrous Separable
     Convolution for Semantic Image Segmentation"
 
     Args:
@@ -180,9 +178,7 @@ class DeepLabV3Plus(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **DeepLabV3Plus**
 
-    Reference:
-        https://arxiv.org/abs/1802.02611v3
-
+    .. __: https://arxiv.org/abs/1802.02611v3
     """
 
     @supports_config_loading

segmentation_models_pytorch/decoders/dpt/model.py

@@ -15,7 +15,7 @@
 
 class DPT(SegmentationModel):
     """
-    DPT is a dense prediction architecture that leverages vision transformers in place of convolutional networks as
+    `DPT`__ is a dense prediction architecture that leverages vision transformers in place of convolutional networks as
     a backbone for dense prediction tasks
 
     It assembles tokens from various stages of the vision transformer into image-like representations at various resolutions
@@ -69,6 +69,7 @@ class DPT(SegmentationModel):
     Returns:
         ``torch.nn.Module``: DPT
 
+    .. __: https://arxiv.org/abs/2103.13413
     """
 
     # fails for encoders with prefix tokens

segmentation_models_pytorch/decoders/fpn/model.py

@@ -12,7 +12,7 @@
 
 
 class FPN(SegmentationModel):
-    """FPN_ is a fully convolution neural network for image semantic segmentation.
+    """`FPN`__ is a fully convolution neural network for image semantic segmentation.
 
     Args:
         encoder_name: Name of the classification model that will be used as an encoder (a.k.a backbone)
@@ -51,9 +51,7 @@ class FPN(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **FPN**
 
-    .. _FPN:
-        http://presentations.cocodataset.org/COCO17-Stuff-FAIR.pdf
-
+    .. __: http://presentations.cocodataset.org/COCO17-Stuff-FAIR.pdf
     """
 
     @supports_config_loading

segmentation_models_pytorch/decoders/linknet/model.py

@@ -13,7 +13,7 @@
 
 
 class Linknet(SegmentationModel):
-    """Linknet_ is a fully convolution neural network for image semantic segmentation. Consist of *encoder*
+    """`Linknet`__ is a fully convolution neural network for image semantic segmentation. Consist of *encoder*
     and *decoder* parts connected with *skip connections*. Encoder extract features of different spatial
     resolution (skip connections) which are used by decoder to define accurate segmentation mask. Use *sum*
     for fusing decoder blocks with skip connections.
@@ -67,8 +67,7 @@ class Linknet(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **Linknet**
 
-    .. _Linknet:
-        https://arxiv.org/abs/1707.03718
+    .. __: https://arxiv.org/abs/1707.03718
     """
 
     @supports_config_loading

segmentation_models_pytorch/decoders/manet/model.py

@@ -13,7 +13,7 @@
 
 
 class MAnet(SegmentationModel):
-    """MAnet_ :  Multi-scale Attention Net. The MA-Net can capture rich contextual dependencies based on
+    """`MAnet`__:  Multi-scale Attention Net. The MA-Net can capture rich contextual dependencies based on
     the attention mechanism, using two blocks:
 
     - Position-wise Attention Block (PAB), which captures the spatial dependencies between pixels in a global view
@@ -72,9 +72,7 @@ class MAnet(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **MAnet**
 
-    .. _MAnet:
-        https://ieeexplore.ieee.org/abstract/document/9201310
-
+    .. __: https://ieeexplore.ieee.org/abstract/document/9201310
     """
 
     @supports_config_loading

segmentation_models_pytorch/decoders/pan/model.py

@@ -13,7 +13,7 @@
 
 
 class PAN(SegmentationModel):
-    """Implementation of PAN_ (Pyramid Attention Network).
+    """Implementation of `PAN`__ (Pyramid Attention Network).
 
     Note:
         Currently works with shape of input tensor >= [B x C x 128 x 128] for pytorch <= 1.1.0
@@ -54,9 +54,7 @@ class PAN(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **PAN**
 
-    .. _PAN:
-        https://arxiv.org/abs/1805.10180
-
+    .. __: https://arxiv.org/abs/1805.10180
     """
 
     @supports_config_loading

segmentation_models_pytorch/decoders/pspnet/model.py

@@ -13,7 +13,7 @@
 
 
 class PSPNet(SegmentationModel):
-    """PSPNet_ is a fully convolution neural network for image semantic segmentation. Consist of
+    """`PSPNet`__ is a fully convolution neural network for image semantic segmentation. Consist of
     *encoder* and *Spatial Pyramid* (decoder). Spatial Pyramid build on top of encoder and does not
     use "fine-features" (features of high spatial resolution). PSPNet can be used for multiclass segmentation
     of high resolution images, however it is not good for detecting small objects and producing accurate,
@@ -68,8 +68,7 @@ class PSPNet(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **PSPNet**
 
-    .. _PSPNet:
-        https://arxiv.org/abs/1612.01105
+    .. __: https://arxiv.org/abs/1612.01105
     """
 
     @supports_config_loading

segmentation_models_pytorch/decoders/segformer/model.py

@@ -12,7 +12,7 @@
 
 
 class Segformer(SegmentationModel):
-    """Segformer is simple and efficient design for semantic segmentation with Transformers
+    """`Segformer`__ is simple and efficient design for semantic segmentation with Transformers
 
     Args:
         encoder_name: Name of the classification model that will be used as an encoder (a.k.a backbone)
@@ -45,9 +45,7 @@ class Segformer(SegmentationModel):
     Returns:
         ``torch.nn.Module``: **Segformer**
 
-    .. _Segformer:
-        https://arxiv.org/abs/2105.15203
-
+    .. __: https://arxiv.org/abs/2105.15203
     """
 
     @supports_config_loading