Fix doc examples (#15257)

docs/source/model_doc/trocr.mdx

@@ -70,7 +70,8 @@ into a single instance to both extract the input features and decode the predict
 >>> processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-handwritten") 
 >>> model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-handwritten")
 
->>> # load image from the IAM dataset url = "https://fki.tic.heia-fr.ch/static/img/a01-122-02.jpg" 
+>>> # load image from the IAM dataset 
+>>> url = "https://fki.tic.heia-fr.ch/static/img/a01-122-02.jpg" 
 >>> image = Image.open(requests.get(url, stream=True).raw).convert("RGB")
 
 >>> pixel_values = processor(image, return_tensors="pt").pixel_values 

src/transformers/models/vilt/modeling_vilt.py

@@ -42,10 +42,10 @@ from .configuration_vilt import ViltConfig
 logger = logging.get_logger(__name__)
 
 _CONFIG_FOR_DOC = "ViltConfig"
-_CHECKPOINT_FOR_DOC = "dandelin/vilt-b32-mlm-itm"
+_CHECKPOINT_FOR_DOC = "dandelin/vilt-b32-mlm"
 
 VILT_PRETRAINED_MODEL_ARCHIVE_LIST = [
-    "dandelin/vilt-b32-mlm-itm",
+    "dandelin/vilt-b32-mlm",
     # See all ViLT models at https://huggingface.co/models?filter=vilt
 ]
 
@@ -775,17 +775,19 @@ class ViltModel(ViltPreTrainedModel):
         Examples:
 
         ```python
-        >>> from transformers import ViltFeatureExtractor, ViltModel
+        >>> from transformers import ViltProcessor, ViltModel
         >>> from PIL import Image
         >>> import requests
 
+        >>> # prepare image and text
         >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
         >>> image = Image.open(requests.get(url, stream=True).raw)
+        >>> text = "hello world"
 
-        >>> feature_extractor = ViltFeatureExtractor.from_pretrained("dandelin/vilt-b32-mlm-itm")
+        >>> processor = ViltProcessor.from_pretrained("dandelin/vilt-b32-mlm")
-        >>> model = ViltModel.from_pretrained("dandelin/vilt-b32-mlm-itm")
+        >>> model = ViltModel.from_pretrained("dandelin/vilt-b32-mlm")
 
-        >>> inputs = feature_extractor(images=image, return_tensors="pt")
+        >>> inputs = processor(image, text, return_tensors="pt")
         >>> outputs = model(**inputs)
         >>> last_hidden_states = outputs.last_hidden_state
         ```"""
@@ -930,10 +932,11 @@ class ViltForMaskedLM(ViltPreTrainedModel):
         >>> from transformers import ViltProcessor, ViltForMaskedLM
         >>> import requests
         >>> from PIL import Image
+        >>> import re
 
         >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
         >>> image = Image.open(requests.get(url, stream=True).raw)
-        >>> text = "How many cats are there?"
+        >>> text = "a bunch of [MASK] laying on a [MASK]."
 
         >>> processor = ViltProcessor.from_pretrained("dandelin/vilt-b32-mlm")
         >>> model = ViltForMaskedLM.from_pretrained("dandelin/vilt-b32-mlm")
@@ -943,7 +946,31 @@ class ViltForMaskedLM(ViltPreTrainedModel):
 
         >>> # forward pass
         >>> outputs = model(**encoding)
-        >>> logits = outputs.logits
+
+        >>> tl = len(re.findall("\[MASK\]", text))
+        >>> inferred_token = [text]
+
+        >>> # gradually fill in the MASK tokens, one by one
+        >>> with torch.no_grad():
+        ...     for i in range(tl):
+        ...         encoded = processor.tokenizer(inferred_token)
+        ...         input_ids = torch.tensor(encoded.input_ids).to(device)
+        ...         encoded = encoded["input_ids"][0][1:-1]
+        ...         outputs = model(input_ids=input_ids, pixel_values=pixel_values)
+        ...         mlm_logits = outputs.logits[0]  # shape (seq_len, vocab_size)
+        ...         # only take into account text features (minus CLS and SEP token)
+        ...         mlm_logits = mlm_logits[1 : input_ids.shape[1] - 1, :]
+        ...         mlm_values, mlm_ids = mlm_logits.softmax(dim=-1).max(dim=-1)
+        ...         # only take into account text
+        ...         mlm_values[torch.tensor(encoded) != 103] = 0
+        ...         select = mlm_values.argmax().item()
+        ...         encoded[select] = mlm_ids[select].item()
+        ...         inferred_token = [processor.decode(encoded)]
+
+        >>> selected_token = ""
+        >>> encoded = processor.tokenizer(inferred_token)
+        >>> processor.decode(encoded.input_ids[0], skip_special_tokens=True)
+        a bunch of cats laying on a couch.
         ```"""
         return_dict = return_dict if return_dict is not None else self.config.use_return_dict
 
@@ -1093,6 +1120,7 @@ class ViltForQuestionAnswering(ViltPreTrainedModel):
         >>> logits = outputs.logits
         >>> idx = logits.argmax(-1).item()
         >>> print("Predicted answer:", model.config.id2label[idx])
+        Predicted answer: 2
         ```"""
         return_dict = return_dict if return_dict is not None else self.config.use_return_dict
 
@@ -1297,13 +1325,13 @@ class ViltForImagesAndTextClassification(ViltPreTrainedModel):
 
         >>> # prepare inputs
         >>> encoding = processor([image1, image2], text, return_tensors="pt")
-        >>> pixel_values = torch.stack([encoding_1.pixel_values, encoding_2.pixel_values], dim=1)
 
         >>> # forward pass
         >>> outputs = model(input_ids=encoding.input_ids, pixel_values=encoding.pixel_values.unsqueeze(0))
         >>> logits = outputs.logits
         >>> idx = logits.argmax(-1).item()
         >>> print("Predicted answer:", model.config.id2label[idx])
+        Predicted answer: True
         ```"""
         output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
         output_hidden_states = (