RegNet
Overview
The RegNet model was proposed in Designing Network Design Spaces by Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollár.
The authors design search spaces to perform Neural Architecture Search (NAS). They first start from a high dimensional search space and iteratively reduce the search space by empirically applying constraints based on the best-performing models sampled by the current search space.
The abstract from the paper is the following:
In this work, we present a new network design paradigm. Our goal is to help advance the understanding of network design and discover design principles that generalize across settings. Instead of focusing on designing individual network instances, we design network design spaces that parametrize populations of networks. The overall process is analogous to classic manual design of networks, but elevated to the design space level. Using our methodology we explore the structure aspect of network design and arrive at a low-dimensional design space consisting of simple, regular networks that we call RegNet. The core insight of the RegNet parametrization is surprisingly simple: widths and depths of good networks can be explained by a quantized linear function. We analyze the RegNet design space and arrive at interesting findings that do not match the current practice of network design. The RegNet design space provides simple and fast networks that work well across a wide range of flop regimes. Under comparable training settings and flops, the RegNet models outperform the popular EfficientNet models while being up to 5x faster on GPUs.
This model was contributed by Francesco. The TensorFlow version of the model was contributed by sayakpaul and ariG23498. The original code can be found here.
The huge 10B model from Self-supervised Pretraining of Visual Features in the Wild, trained on one billion Instagram images, is available on the hub
Resources
A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with RegNet.
- RegNetForImageClassification is supported by this example script and notebook.
- See also: Image classification task guide
If you’re interested in submitting a resource to be included here, please feel free to open a Pull Request and we’ll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
RegNetConfig
class transformers.RegNetConfig
< source >( num_channels = 3 embedding_size = 32 hidden_sizes = [128, 192, 512, 1088] depths = [2, 6, 12, 2] groups_width = 64 layer_type = 'y' hidden_act = 'relu' **kwargs )
Parameters
- num_channels (
int, optional, defaults to 3) — The number of input channels. - embedding_size (
int, optional, defaults to 64) — Dimensionality (hidden size) for the embedding layer. - hidden_sizes (
List[int], optional, defaults to[256, 512, 1024, 2048]) — Dimensionality (hidden size) at each stage. - depths (
List[int], optional, defaults to[3, 4, 6, 3]) — Depth (number of layers) for each stage. - layer_type (
str, optional, defaults to"y") — The layer to use, it can be either"x" or“y”. Anxlayer is a ResNet's BottleNeck layer withreductionfixed to1. While aylayer is ax` but with squeeze and excitation. Please refer to the paper for a detailed explanation of how these layers were constructed. - hidden_act (
str, optional, defaults to"relu") — The non-linear activation function in each block. If string,"gelu","relu","selu"and"gelu_new"are supported. - downsample_in_first_stage (
bool, optional, defaults toFalse) — IfTrue, the first stage will downsample the inputs using astrideof 2.
This is the configuration class to store the configuration of a RegNetModel. It is used to instantiate a RegNet model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the RegNet facebook/regnet-y-040 architecture.
Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.
Example:
>>> from transformers import RegNetConfig, RegNetModel
>>> # Initializing a RegNet regnet-y-40 style configuration
>>> configuration = RegNetConfig()
>>> # Initializing a model from the regnet-y-40 style configuration
>>> model = RegNetModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.configRegNetModel
class transformers.RegNetModel
< source >( config )
Parameters
- config (RegNetConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
The bare RegNet model outputting raw features without any specific head on top. This model is a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matters related to general usage and behavior.
forward
< source >( pixel_values: Tensor output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_outputs.BaseModelOutputWithPoolingAndNoAttention or tuple(torch.FloatTensor)
Parameters
- pixel_values (
torch.FloatTensorof shape(batch_size, num_channels, height, width)) — Pixel values. Pixel values can be obtained using AutoImageProcessor. See ConvNextImageProcessor.call() for details. - output_hidden_states (
bool, optional) — Whether or not to return the hidden states of all layers. Seehidden_statesunder returned tensors for more detail. - return_dict (
bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_outputs.BaseModelOutputWithPoolingAndNoAttention or tuple(torch.FloatTensor)
A transformers.modeling_outputs.BaseModelOutputWithPoolingAndNoAttention or a tuple of
torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various
elements depending on the configuration (RegNetConfig) and inputs.
-
last_hidden_state (
torch.FloatTensorof shape(batch_size, num_channels, height, width)) — Sequence of hidden-states at the output of the last layer of the model. -
pooler_output (
torch.FloatTensorof shape(batch_size, hidden_size)) — Last layer hidden-state after a pooling operation on the spatial dimensions. -
hidden_states (
tuple(torch.FloatTensor), optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.FloatTensor(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, num_channels, height, width).Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
The RegNetModel forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoImageProcessor, RegNetModel
>>> import torch
>>> from datasets import load_dataset
>>> dataset = load_dataset("huggingface/cats-image")
>>> image = dataset["test"]["image"][0]
>>> image_processor = AutoImageProcessor.from_pretrained("facebook/regnet-y-040")
>>> model = RegNetModel.from_pretrained("facebook/regnet-y-040")
>>> inputs = image_processor(image, return_tensors="pt")
>>> with torch.no_grad():
... outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_state
>>> list(last_hidden_states.shape)
[1, 1088, 7, 7]RegNetForImageClassification
class transformers.RegNetForImageClassification
< source >( config )
Parameters
- config (RegNetConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
RegNet Model with an image classification head on top (a linear layer on top of the pooled features), e.g. for ImageNet.
This model is a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matters related to general usage and behavior.
forward
< source >( pixel_values: Optional = None labels: Optional = None output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_outputs.ImageClassifierOutputWithNoAttention or tuple(torch.FloatTensor)
Parameters
- pixel_values (
torch.FloatTensorof shape(batch_size, num_channels, height, width)) — Pixel values. Pixel values can be obtained using AutoImageProcessor. See ConvNextImageProcessor.call() for details. - output_hidden_states (
bool, optional) — Whether or not to return the hidden states of all layers. Seehidden_statesunder returned tensors for more detail. - return_dict (
bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - labels (
torch.LongTensorof shape(batch_size,), optional) — Labels for computing the image classification/regression loss. Indices should be in[0, ..., config.num_labels - 1]. Ifconfig.num_labels > 1a classification loss is computed (Cross-Entropy).
Returns
transformers.modeling_outputs.ImageClassifierOutputWithNoAttention or tuple(torch.FloatTensor)
A transformers.modeling_outputs.ImageClassifierOutputWithNoAttention or a tuple of
torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various
elements depending on the configuration (RegNetConfig) and inputs.
- loss (
torch.FloatTensorof shape(1,), optional, returned whenlabelsis provided) — Classification (or regression if config.num_labels==1) loss. - logits (
torch.FloatTensorof shape(batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (before SoftMax). - hidden_states (
tuple(torch.FloatTensor), optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftorch.FloatTensor(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each stage) of shape(batch_size, num_channels, height, width). Hidden-states (also called feature maps) of the model at the output of each stage.
The RegNetForImageClassification forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoImageProcessor, RegNetForImageClassification
>>> import torch
>>> from datasets import load_dataset
>>> dataset = load_dataset("huggingface/cats-image")
>>> image = dataset["test"]["image"][0]
>>> image_processor = AutoImageProcessor.from_pretrained("facebook/regnet-y-040")
>>> model = RegNetForImageClassification.from_pretrained("facebook/regnet-y-040")
>>> inputs = image_processor(image, return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> # model predicts one of the 1000 ImageNet classes
>>> predicted_label = logits.argmax(-1).item()
>>> print(model.config.id2label[predicted_label])
tabby, tabby catTFRegNetModel
class transformers.TFRegNetModel
< source >( config: RegNetConfig *inputs **kwargs )
Parameters
- config (RegNetConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
The bare RegNet model outputting raw features without any specific head on top. This model is a Tensorflow keras.layers.Layer sub-class. Use it as a regular Tensorflow Module and refer to the Tensorflow documentation for all matter related to general usage and behavior.
call
< source >( pixel_values: Tensor output_hidden_states: Optional = None return_dict: Optional = None training: bool = False ) → transformers.modeling_tf_outputs.TFBaseModelOutputWithPoolingAndNoAttention or tuple(tf.Tensor)
Parameters
- pixel_values (
tf.Tensorof shape(batch_size, num_channels, height, width)) — Pixel values. Pixel values can be obtained using AutoImageProcessor. SeeConveNextImageProcessor.__call__for details. - output_hidden_states (
bool, optional) — Whether or not to return the hidden states of all layers. Seehidden_statesunder returned tensors for more detail. - return_dict (
bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
Returns
transformers.modeling_tf_outputs.TFBaseModelOutputWithPoolingAndNoAttention or tuple(tf.Tensor)
A transformers.modeling_tf_outputs.TFBaseModelOutputWithPoolingAndNoAttention or a tuple of tf.Tensor (if
return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the
configuration (RegNetConfig) and inputs.
-
last_hidden_state (
tf.Tensorof shape(batch_size, num_channels, height, width)) — Sequence of hidden-states at the output of the last layer of the model. -
pooler_output (
tf.Tensorof shape(batch_size, hidden_size)) — Last layer hidden-state after a pooling operation on the spatial dimensions. -
hidden_states (
tuple(tf.Tensor), optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftf.Tensor(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape(batch_size, num_channels, height, width).Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
The TFRegNetModel forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoImageProcessor, TFRegNetModel
>>> from datasets import load_dataset
>>> dataset = load_dataset("huggingface/cats-image")
>>> image = dataset["test"]["image"][0]
>>> image_processor = AutoImageProcessor.from_pretrained("facebook/regnet-y-040")
>>> model = TFRegNetModel.from_pretrained("facebook/regnet-y-040")
>>> inputs = image_processor(image, return_tensors="tf")
>>> outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_state
>>> list(last_hidden_states.shape)
[1, 1088, 7, 7]TFRegNetForImageClassification
class transformers.TFRegNetForImageClassification
< source >( config: RegNetConfig *inputs **kwargs )
Parameters
- config (RegNetConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
RegNet Model with an image classification head on top (a linear layer on top of the pooled features), e.g. for ImageNet.
This model is a Tensorflow keras.layers.Layer sub-class. Use it as a regular Tensorflow Module and refer to the Tensorflow documentation for all matter related to general usage and behavior.
call
< source >( pixel_values: Optional = None labels: Optional = None output_hidden_states: Optional = None return_dict: Optional = None training: bool = False ) → transformers.modeling_tf_outputs.TFSequenceClassifierOutput or tuple(tf.Tensor)
Parameters
- pixel_values (
tf.Tensorof shape(batch_size, num_channels, height, width)) — Pixel values. Pixel values can be obtained using AutoImageProcessor. SeeConveNextImageProcessor.__call__for details. - output_hidden_states (
bool, optional) — Whether or not to return the hidden states of all layers. Seehidden_statesunder returned tensors for more detail. - return_dict (
bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple. - labels (
tf.Tensorof shape(batch_size,), optional) — Labels for computing the image classification/regression loss. Indices should be in[0, ..., config.num_labels - 1]. Ifconfig.num_labels > 1a classification loss is computed (Cross-Entropy).
Returns
transformers.modeling_tf_outputs.TFSequenceClassifierOutput or tuple(tf.Tensor)
A transformers.modeling_tf_outputs.TFSequenceClassifierOutput or a tuple of tf.Tensor (if
return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the
configuration (RegNetConfig) and inputs.
-
loss (
tf.Tensorof shape(batch_size, ), optional, returned whenlabelsis provided) — Classification (or regression if config.num_labels==1) loss. -
logits (
tf.Tensorof shape(batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (before SoftMax). -
hidden_states (
tuple(tf.Tensor), optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple oftf.Tensor(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size).Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(tf.Tensor), optional, returned whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple oftf.Tensor(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length).Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
The TFRegNetForImageClassification forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoImageProcessor, TFRegNetForImageClassification
>>> import tensorflow as tf
>>> from datasets import load_dataset
>>> dataset = load_dataset("huggingface/cats-image")
>>> image = dataset["test"]["image"][0]
>>> image_processor = AutoImageProcessor.from_pretrained("facebook/regnet-y-040")
>>> model = TFRegNetForImageClassification.from_pretrained("facebook/regnet-y-040")
>>> inputs = image_processor(image, return_tensors="tf")
>>> logits = model(**inputs).logits
>>> # model predicts one of the 1000 ImageNet classes
>>> predicted_label = int(tf.math.argmax(logits, axis=-1))
>>> print(model.config.id2label[predicted_label])
tabby, tabby catFlaxRegNetModel
class transformers.FlaxRegNetModel
< source >( config: RegNetConfig input_shape = (1, 224, 224, 3) seed: int = 0 dtype: dtype = <class 'jax.numpy.float32'> _do_init: bool = True **kwargs )
Parameters
- config (RegNetConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
- dtype (
jax.numpy.dtype, optional, defaults tojax.numpy.float32) — The data type of the computation. Can be one ofjax.numpy.float32,jax.numpy.float16(on GPUs) andjax.numpy.bfloat16(on TPUs).This can be used to enable mixed-precision training or half-precision inference on GPUs or TPUs. If specified all the computation will be performed with the given
dtype.Note that this only specifies the dtype of the computation and does not influence the dtype of model parameters.
If you wish to change the dtype of the model parameters, see to_fp16() and to_bf16().
The bare RegNet model outputting raw features without any specific head on top.
This model inherits from FlaxPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)
This model is also a flax.linen.Module subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.
Finally, this model supports inherent JAX features such as:
__call__
< source >( pixel_values params: dict = None train: bool = False output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling or tuple(torch.FloatTensor)
Returns
transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxBaseModelOutputWithPooling or a tuple of
torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various
elements depending on the configuration (<class 'transformers.models.regnet.configuration_regnet.RegNetConfig'>) and inputs.
-
last_hidden_state (
jnp.ndarrayof shape(batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model. -
pooler_output (
jnp.ndarrayof shape(batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) further processed by a Linear layer and a Tanh activation function. The Linear layer weights are trained from the next sentence prediction (classification) objective during pretraining. -
hidden_states (
tuple(jnp.ndarray), optional, returned whenoutput_hidden_states=Trueis passed or whenconfig.output_hidden_states=True) — Tuple ofjnp.ndarray(one for the output of the embeddings + one for the output of each layer) of shape(batch_size, sequence_length, hidden_size).Hidden-states of the model at the output of each layer plus the initial embedding outputs.
-
attentions (
tuple(jnp.ndarray), optional, returned whenoutput_attentions=Trueis passed or whenconfig.output_attentions=True) — Tuple ofjnp.ndarray(one for each layer) of shape(batch_size, num_heads, sequence_length, sequence_length).Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
The FlaxRegNetPreTrainedModel forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Examples:
>>> from transformers import AutoImageProcessor, FlaxRegNetModel
>>> from PIL import Image
>>> import requests
>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> image_processor = AutoImageProcessor.from_pretrained("facebook/regnet-y-040")
>>> model = FlaxRegNetModel.from_pretrained("facebook/regnet-y-040")
>>> inputs = image_processor(images=image, return_tensors="np")
>>> outputs = model(**inputs)
>>> last_hidden_states = outputs.last_hidden_stateFlaxRegNetForImageClassification
class transformers.FlaxRegNetForImageClassification
< source >( config: RegNetConfig input_shape = (1, 224, 224, 3) seed: int = 0 dtype: dtype = <class 'jax.numpy.float32'> _do_init: bool = True **kwargs )
Parameters
- config (RegNetConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.
- dtype (
jax.numpy.dtype, optional, defaults tojax.numpy.float32) — The data type of the computation. Can be one ofjax.numpy.float32,jax.numpy.float16(on GPUs) andjax.numpy.bfloat16(on TPUs).This can be used to enable mixed-precision training or half-precision inference on GPUs or TPUs. If specified all the computation will be performed with the given
dtype.Note that this only specifies the dtype of the computation and does not influence the dtype of model parameters.
If you wish to change the dtype of the model parameters, see to_fp16() and to_bf16().
RegNet Model with an image classification head on top (a linear layer on top of the pooled features), e.g. for ImageNet.
This model inherits from FlaxPreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading, saving and converting weights from PyTorch models)
This model is also a flax.linen.Module subclass. Use it as a regular Flax linen Module and refer to the Flax documentation for all matter related to general usage and behavior.
Finally, this model supports inherent JAX features such as:
__call__
< source >( pixel_values params: dict = None train: bool = False output_hidden_states: Optional = None return_dict: Optional = None ) → transformers.modeling_flax_outputs.FlaxImageClassifierOutputWithNoAttention or tuple(torch.FloatTensor)
Returns
transformers.modeling_flax_outputs.FlaxImageClassifierOutputWithNoAttention or tuple(torch.FloatTensor)
A transformers.modeling_flax_outputs.FlaxImageClassifierOutputWithNoAttention or a tuple of
torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various
elements depending on the configuration (<class 'transformers.models.regnet.configuration_regnet.RegNetConfig'>) and inputs.
- logits (
jnp.ndarrayof shape(batch_size, config.num_labels)) — Classification (or regression if config.num_labels==1) scores (before SoftMax). - hidden_states (
tuple(jnp.ndarray), optional, returned whenoutput_hidden_states=Trueis passed or when config.output_hidden_states=True): Tuple ofjnp.ndarray(one for the output of the embeddings, if the model has an embedding layer, + one for the output of each stage) of shape(batch_size, num_channels, height, width). Hidden-states (also called feature maps) of the model at the output of each stage.
The FlaxRegNetPreTrainedModel forward method, overrides the __call__ special method.
Although the recipe for forward pass needs to be defined within this function, one should call the Module
instance afterwards instead of this since the former takes care of running the pre and post processing steps while
the latter silently ignores them.
Example:
>>> from transformers import AutoImageProcessor, FlaxRegNetForImageClassification
>>> from PIL import Image
>>> import jax
>>> import requests
>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)
>>> image_processor = AutoImageProcessor.from_pretrained("facebook/regnet-y-040")
>>> model = FlaxRegNetForImageClassification.from_pretrained("facebook/regnet-y-040")
>>> inputs = image_processor(images=image, return_tensors="np")
>>> outputs = model(**inputs)
>>> logits = outputs.logits
>>> # model predicts one of the 1000 ImageNet classes
>>> predicted_class_idx = jax.numpy.argmax(logits, axis=-1)
>>> print("Predicted class:", model.config.id2label[predicted_class_idx.item()])