Map model_type and doc pages names (#14944)

* Map model_type and doc pages names

* Add script

* Fix typo

* Quality

* Manual check for Auto
Co-authored-by: Lysandre <lysandre.debut@reseau.eseo.fr>

docs/source/model_doc/trocr.mdx

@@ -48,7 +48,7 @@ Tips:
   on both printed (e.g. the [SROIE dataset](https://paperswithcode.com/dataset/sroie) and handwritten (e.g. the [IAM
   Handwriting dataset](https://fki.tic.heia-fr.ch/databases/iam-handwriting-database>) text recognition tasks. For more
   information, see the [official models](https://huggingface.co/models?other=trocr>).
-- TrOCR is always used within the [VisionEncoderDecoder](./model_doc/visionencoderdecoder) framework.
+- TrOCR is always used within the [VisionEncoderDecoder](vision-encoder-decoder) framework.
 
 ## Inference
 

docs/source/model_summary.mdx

@@ -66,7 +66,7 @@ that at each position, the model can only look at the tokens before the attentio
 <a href="https://huggingface.co/models?filter=openai-gpt">
 <img alt="Models" src="https://img.shields.io/badge/All_model_pages-openai--gpt-blueviolet">
 </a>
-<a href="model_doc/gpt">
+<a href="model_doc/openai-gpt">
 <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-openai--gpt-blueviolet">
 </a>
 
@@ -118,7 +118,7 @@ The library provides a version of the model for language modeling only.
 <a href="https://huggingface.co/models?filter=transfo-xl">
 <img alt="Models" src="https://img.shields.io/badge/All_model_pages-transfo--xl-blueviolet">
 </a>
-<a href="model_doc/transformerxl">
+<a href="model_doc/transfo-xl">
 <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-transfo--xl-blueviolet">
 </a>
 
@@ -367,7 +367,7 @@ question answering.
 <a href="https://huggingface.co/models?filter=xlm-roberta">
 <img alt="Models" src="https://img.shields.io/badge/All_model_pages-xlm--roberta-blueviolet">
 </a>
-<a href="model_doc/xlmroberta">
+<a href="model_doc/xlm-roberta">
 <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-xlm--roberta-blueviolet">
 </a>
 
@@ -649,7 +649,7 @@ summarization.
 <a href="https://huggingface.co/models?filter=xprophetnet">
 <img alt="Models" src="https://img.shields.io/badge/All_model_pages-xprophetnet-blueviolet">
 </a>
-<a href="model_doc/xlmprophetnet">
+<a href="model_doc/xlm-prophetnet">
 <img alt="Doc" src="https://img.shields.io/badge/Model_documentation-xprophetnet-blueviolet">
 </a>
 

src/transformers/models/auto/configuration_auto.py

@@ -279,6 +279,16 @@ MODEL_NAMES_MAPPING = OrderedDict(
         ("unispeech-sat", "UniSpeechSat"),
         ("unispeech", "UniSpeech"),
         ("wavlm", "WavLM"),
+        ("bort", "BORT"),
+        ("dialogpt", "DialoGPT"),
+        ("xls_r", "XLS-R"),
+        ("t5v1.1", "T5v1.1"),
+        ("herbert", "HerBERT"),
+        ("wav2vec2_phoneme", "Wav2Vec2Phoneme"),
+        ("megatron_gpt2", "MegatronGPT2"),
+        ("xlsr_wav2vec2", "XLSR-Wav2Vec2"),
+        ("mluke", "mLUKE"),
+        ("layoutxlm", "LayoutXLM"),
     ]
 )
 

utils/check_repo.py

@@ -18,6 +18,7 @@ import inspect
 import os
 import re
 import warnings
+from difflib import get_close_matches
 from pathlib import Path
 
 from transformers import is_flax_available, is_tf_available, is_torch_available
@@ -596,6 +597,33 @@ def check_all_objects_are_documented():
             + "\n - ".join(undocumented_objs)
         )
     check_docstrings_are_in_md()
+    check_model_type_doc_match()
+
+
+def check_model_type_doc_match():
+    """Check all doc pages have a corresponding model type."""
+    model_doc_folder = Path(PATH_TO_DOC) / "model_doc"
+    model_docs = [m.stem for m in model_doc_folder.glob("*.mdx")]
+
+    model_types = list(transformers.models.auto.configuration_auto.MODEL_NAMES_MAPPING.keys())
+
+    errors = []
+    for m in model_docs:
+        if m not in model_types and m != "auto":
+            close_matches = get_close_matches(m, model_types)
+            error_message = f"{m} is not a proper model identifier."
+            if len(close_matches) > 0:
+                close_matches = "/".join(close_matches)
+                error_message += f" Did you mean {close_matches}?"
+            errors.append(error_message)
+
+    if len(errors) > 0:
+        raise ValueError(
+            "Some model doc pages do not match any existing model type:\n"
+            + "\n".join(errors)
+            + "\nYou can add any missing model type to the `MODEL_NAMES_MAPPING` constant in "
+            "models/auto/configuration_auto.py."
+        )
 
 
 # Re pattern to catch :obj:`xx`, :class:`xx`, :func:`xx` or :meth:`xx`.