Update ALIGN docs (#22025)

* Fix typos and add code examples, resources

docs/source/en/model_doc/align.mdx

@@ -14,15 +14,57 @@ specific language governing permissions and limitations under the License.
 
 ## Overview
 
-The ALIGN model was proposed in [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://arxiv.org/abs/2102.05918) by Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc V. Le, Yunhsuan Sung, Zhen Li, Tom Duerig. ALIGN features a dual-encoder architecture with [EfficientNet](efficientnet) as its vision encoder and [BERT](bert) as its text encoder, and learns to align visual and text representations with contrastive learning. Unlike previous work, ALIGN leverages a massive noisy dataset and shows that the scale of the corpus can be used to achieve SOTA representations with a simple recipe.
+The ALIGN model was proposed in [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://arxiv.org/abs/2102.05918) by Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc V. Le, Yunhsuan Sung, Zhen Li, Tom Duerig. ALIGN is a multi-modal vision and language model. It can be used for image-text similarity and for zero-shot image classification. ALIGN features a dual-encoder architecture with [EfficientNet](efficientnet) as its vision encoder and [BERT](bert) as its text encoder, and learns to align visual and text representations with contrastive learning. Unlike previous work, ALIGN leverages a massive noisy dataset and shows that the scale of the corpus can be used to achieve SOTA representations with a simple recipe.
 
 The abstract from the paper is the following:
 
 *Pre-trained representations are becoming crucial for many NLP and perception tasks. While representation learning in NLP has transitioned to training on raw text without human annotations, visual and vision-language representations still rely heavily on curated training datasets that are expensive or require expert knowledge. For vision applications, representations are mostly learned using datasets with explicit class labels such as ImageNet or OpenImages. For vision-language, popular datasets like Conceptual Captions, MSCOCO, or CLIP all involve a non-trivial data collection (and cleaning) process. This costly curation process limits the size of datasets and hence hinders the scaling of trained models. In this paper, we leverage a noisy dataset of over one billion image alt-text pairs, obtained without expensive filtering or post-processing steps in the Conceptual Captions dataset. A simple dual-encoder architecture learns to align visual and language representations of the image and text pairs using a contrastive loss. We show that the scale of our corpus can make up for its noise and leads to state-of-the-art representations even with such a simple learning scheme. Our visual representation achieves strong performance when transferred to classification tasks such as ImageNet and VTAB. The aligned visual and language representations enables zero-shot image classification and also set new state-of-the-art results on Flickr30K and MSCOCO image-text retrieval benchmarks, even when compared with more sophisticated cross-attention models. The representations also enable cross-modality search with complex text and text + image queries.*
 
+## Usage
+
+ALIGN uses EfficientNet to get visual features and BERT to get the text features. Both the text and visual features are then projected to a latent space with identical dimension. The dot product between the projected image and text features is then used as a similarity score.
+
+[`AlignProcessor`] wraps [`EfficientNetImageProcessor`] and [`BertTokenizer`] into a single instance to both encode the text and preprocess the images. The following example shows how to get the image-text similarity scores using [`AlignProcessor`] and [`AlignModel`].
+
+```python
+import requests
+import torch
+from PIL import Image
+from transformers import AlignProcessor, AlignModel
+
+processor = AlignProcessor.from_pretrained("kakaobrain/align-base")
+model = AlignModel.from_pretrained("kakaobrain/align-base")
+
+url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+image = Image.open(requests.get(url, stream=True).raw)
+candidate_labels = ["an image of a cat", "an image of a dog"]
+
+inputs = processor(text=candidate_labels, images=image, return_tensors="pt")
+
+with torch.no_grad():
+    outputs = model(**inputs)
+
+# this is the image-text similarity score
+logits_per_image = outputs.logits_per_image
+
+# we can take the softmax to get the label probabilities
+probs = logits_per_image.softmax(dim=1)
+print(probs)
+```
+
 This model was contributed by [Alara Dirik](https://huggingface.co/adirik).
 The original code is not released, this implementation is based on the Kakao Brain implementation based on the original paper.
 
+## Resources
+
+A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with ALIGN.
+
+- A blog post on [ALIGN and the COYO-700M dataset](https://huggingface.co/blog).
+- A zero-shot image classification [demo](https://huggingface.co/spaces/adirik/ALIGN-zero-shot-image-classification).
+- [Model card](https://huggingface.co/kakaobrain/align-base) of `kakaobrain/align-base` model.
+
+If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we will review it. The resource should ideally demonstrate something new instead of duplicating an existing resource.
+
 
 ## AlignConfig
 

src/transformers/models/align/modeling_align.py

@@ -122,7 +122,7 @@ ALIGN_VISION_INPUTS_DOCSTRING = r"""
     Args:
         pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):
             Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using
-            [`AutoImageProcessor`]. See [`CLIPImageProcessor.__call__`] for details.
+            [`AutoImageProcessor`]. See [`EfficientNetImageProcessor.__call__`] for details.
         output_hidden_states (`bool`, *optional*):
             Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
             more detail.
@@ -172,7 +172,7 @@ ALIGN_INPUTS_DOCSTRING = r"""
             model's internal embedding lookup matrix.
         pixel_values (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):
             Pixel values. Padding will be ignored by default should you provide it. Pixel values can be obtained using
-            [`AutoImageProcessor`]. See [`CLIPImageProcessor.__call__`] for details.
+            [`AutoImageProcessor`]. See [`EfficientNetImageProcessor.__call__`] for details.
         return_loss (`bool`, *optional*):
             Whether or not to return the contrastive loss.
         output_attentions (`bool`, *optional*):

src/transformers/models/align/processing_align.py

@@ -31,7 +31,7 @@ class AlignProcessor(ProcessorMixin):
     Args:
         image_processor ([`EfficientNetImageProcessor`]):
             The image processor is a required input.
-        tokenizer ([`BERTTokenizer`, `BertTokenizerFast`]):
+        tokenizer ([`BertTokenizer`, `BertTokenizerFast`]):
             The tokenizer is a required input.
     """