Convert rst files (#14888)

* Convert all tutorials and guides

* Convert all remaining rst to mdx

* Track and fix bad links

src/transformers/models/splinter/modeling_splinter.py

@@ -605,8 +605,8 @@ SPLINTER_INPUTS_DOCSTRING = r"""
 )
 class SplinterModel(SplinterPreTrainedModel):
     """
-    The model is an encoder (with only self-attention) following the architecture described in `Attention is all you
-    need <https://arxiv.org/abs/1706.03762>`__ by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion
+    The model is an encoder (with only self-attention) following the architecture described in [Attention is all you
+    need](https://arxiv.org/abs/1706.03762) by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion
     Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.
     """
 

src/transformers/models/unispeech/modeling_unispeech.py

@@ -129,8 +129,8 @@ def _compute_mask_indices(
     min_masks: int = 0,
 ) -> np.ndarray:
     """
-    Computes random mask spans for a given shape. Used to implement `SpecAugment: A Simple Data Augmentation Method for
-    ASR <https://arxiv.org/abs/1904.08779>`__. Note that this method is not optimized to run on TPU and should be run
+    Computes random mask spans for a given shape. Used to implement [SpecAugment: A Simple Data Augmentation Method for
+    ASR](https://arxiv.org/abs/1904.08779). Note that this method is not optimized to run on TPU and should be run
     on CPU as part of the preprocessing during training.
 
     Args:
@@ -836,8 +836,8 @@ class UniSpeechEncoderStableLayerNorm(nn.Module):
 
 class UniSpeechGumbelVectorQuantizer(nn.Module):
     """
-    Vector quantization using gumbel softmax. See `CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX
-    <https://arxiv.org/pdf/1611.01144.pdf>`__ for more information.
+    Vector quantization using gumbel softmax. See [CATEGORICAL REPARAMETERIZATION WITH
+    GUMBEL-SOFTMAX](https://arxiv.org/pdf/1611.01144.pdf) for more information.
     """
 
     def __init__(self, config):
@@ -1068,8 +1068,8 @@ class UniSpeechModel(UniSpeechPreTrainedModel):
         attention_mask: Optional[torch.LongTensor] = None,
     ):
         """
-        Masks extracted features along time axis and/or along feature axis according to `SpecAugment
-        <https://arxiv.org/abs/1904.08779>`__ .
+        Masks extracted features along time axis and/or along feature axis according to
+        [SpecAugment](https://arxiv.org/abs/1904.08779).
         """
 
         # `config.apply_spec_augment` can set masking to False

src/transformers/models/unispeech_sat/modeling_unispeech_sat.py

@@ -163,8 +163,8 @@ def _compute_mask_indices(
     min_masks: int = 0,
 ) -> np.ndarray:
     """
-    Computes random mask spans for a given shape. Used to implement `SpecAugment: A Simple Data Augmentation Method for
-    ASR <https://arxiv.org/abs/1904.08779>`__. Note that this method is not optimized to run on TPU and should be run
+    Computes random mask spans for a given shape. Used to implement [SpecAugment: A Simple Data Augmentation Method for
+    ASR](https://arxiv.org/abs/1904.08779). Note that this method is not optimized to run on TPU and should be run
     on CPU as part of the preprocessing during training.
 
     Args:
@@ -870,8 +870,8 @@ class UniSpeechSatEncoderStableLayerNorm(nn.Module):
 
 class UniSpeechSatGumbelVectorQuantizer(nn.Module):
     """
-    Vector quantization using gumbel softmax. See `CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX
-    <https://arxiv.org/pdf/1611.01144.pdf>`__ for more information.
+    Vector quantization using gumbel softmax. See [CATEGORICAL REPARAMETERIZATION WITH
+    GUMBEL-SOFTMAX](https://arxiv.org/pdf/1611.01144.pdf) for more information.
     """
 
     def __init__(self, config):
@@ -1101,8 +1101,8 @@ class UniSpeechSatModel(UniSpeechSatPreTrainedModel):
         attention_mask: Optional[torch.LongTensor] = None,
     ):
         """
-        Masks extracted features along time axis and/or along feature axis according to `SpecAugment
-        <https://arxiv.org/abs/1904.08779>`__ .
+        Masks extracted features along time axis and/or along feature axis according to
+        [SpecAugment](https://arxiv.org/abs/1904.08779).
         """
 
         # `config.apply_spec_augment` can set masking to False

src/transformers/models/visual_bert/modeling_visual_bert.py

@@ -679,8 +679,8 @@ VISUAL_BERT_INPUTS_DOCSTRING = r"""
 class VisualBertModel(VisualBertPreTrainedModel):
     """
 
-    The model can behave as an encoder (with only self-attention) following the architecture described in `Attention is
-    all you need <https://arxiv.org/abs/1706.03762>`__ by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit,
+    The model can behave as an encoder (with only self-attention) following the architecture described in [Attention is
+    all you need](https://arxiv.org/abs/1706.03762) by Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit,
     Llion Jones, Aidan N. Gomez, Lukasz Kaiser and Illia Polosukhin.
     """
 

src/transformers/models/wav2vec2/modeling_flax_wav2vec2.py

@@ -114,8 +114,8 @@ def _compute_mask_indices(
     min_masks: int = 0,
 ) -> np.ndarray:
     """
-    Computes random mask spans for a given shape. Used to implement `SpecAugment: A Simple Data Augmentation Method for
-    ASR <https://arxiv.org/abs/1904.08779>`__. Note that this method is not optimized to run on TPU and should be run
+    Computes random mask spans for a given shape. Used to implement [SpecAugment: A Simple Data Augmentation Method for
+    ASR](https://arxiv.org/abs/1904.08779). Note that this method is not optimized to run on TPU and should be run
     on CPU as part of the preprocessing during training.
 
     Args:
@@ -685,8 +685,8 @@ class FlaxWav2Vec2StableLayerNormEncoder(nn.Module):
 
 class FlaxWav2Vec2GumbelVectorQuantizer(nn.Module):
     """
-    Vector quantization using gumbel softmax. See `CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX
-    <https://arxiv.org/pdf/1611.01144.pdf>`__ for more information.
+    Vector quantization using gumbel softmax. See [CATEGORICAL REPARAMETERIZATION WITH
+    GUMBEL-SOFTMAX](https://arxiv.org/pdf/1611.01144.pdf) for more information.
     """
 
     config: Wav2Vec2Config

src/transformers/models/wav2vec2/modeling_tf_wav2vec2.py

@@ -214,8 +214,8 @@ def _compute_mask_indices(
         mask_length: size of the mask
         min_masks: minimum number of masked spans
 
-    Adapted from `fairseq's data_utils.py
-    <https://github.com/pytorch/fairseq/blob/e0788f7007a8473a76db573985031f3c94201e79/fairseq/data/data_utils.py#L376>`__.
+    Adapted from [fairseq's
+    data_utils.py](https://github.com/pytorch/fairseq/blob/e0788f7007a8473a76db573985031f3c94201e79/fairseq/data/data_utils.py#L376).
     """
     batch_size, sequence_length = shape
 
@@ -1137,8 +1137,8 @@ class TFWav2Vec2MainLayer(tf.keras.layers.Layer):
 
     def _mask_hidden_states(self, hidden_states: tf.Tensor, mask_time_indices: Optional[tf.Tensor] = None):
         """
-        Masks extracted features along time axis and/or along feature axis according to `SpecAugment
-        <https://arxiv.org/abs/1904.08779>`__ .
+        Masks extracted features along time axis and/or along feature axis according to
+        [SpecAugment](https://arxiv.org/abs/1904.08779).
         """
         batch_size, sequence_length, hidden_size = shape_list(hidden_states)
 

src/transformers/models/wav2vec2/modeling_wav2vec2.py

@@ -177,8 +177,8 @@ def _compute_mask_indices(
     min_masks: int = 0,
 ) -> np.ndarray:
     """
-    Computes random mask spans for a given shape. Used to implement `SpecAugment: A Simple Data Augmentation Method for
-    ASR <https://arxiv.org/abs/1904.08779>`__. Note that this method is not optimized to run on TPU and should be run
+    Computes random mask spans for a given shape. Used to implement [SpecAugment: A Simple Data Augmentation Method for
+    ASR](https://arxiv.org/abs/1904.08779). Note that this method is not optimized to run on TPU and should be run
     on CPU as part of the preprocessing during training.
 
     Args:
@@ -907,8 +907,8 @@ class Wav2Vec2EncoderStableLayerNorm(nn.Module):
 
 class Wav2Vec2GumbelVectorQuantizer(nn.Module):
     """
-    Vector quantization using gumbel softmax. See `CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX
-    <https://arxiv.org/pdf/1611.01144.pdf>`__ for more information.
+    Vector quantization using gumbel softmax. See `[CATEGORICAL REPARAMETERIZATION WITH
+    GUMBEL-SOFTMAX](https://arxiv.org/pdf/1611.01144.pdf) for more information.
     """
 
     def __init__(self, config):
@@ -1212,8 +1212,8 @@ class Wav2Vec2Model(Wav2Vec2PreTrainedModel):
         attention_mask: Optional[torch.LongTensor] = None,
     ):
         """
-        Masks extracted features along time axis and/or along feature axis according to `SpecAugment
-        <https://arxiv.org/abs/1904.08779>`__ .
+        Masks extracted features along time axis and/or along feature axis according to
+        [SpecAugment](https://arxiv.org/abs/1904.08779).
         """
 
         # `config.apply_spec_augment` can set masking to False

src/transformers/models/wavlm/modeling_wavlm.py

@@ -130,8 +130,8 @@ def _compute_mask_indices(
     min_masks: int = 0,
 ) -> np.ndarray:
     """
-    Computes random mask spans for a given shape. Used to implement `SpecAugment: A Simple Data Augmentation Method for
-    ASR <https://arxiv.org/abs/1904.08779>`__. Note that this method is not optimized to run on TPU and should be run
+    Computes random mask spans for a given shape. Used to implement [SpecAugment: A Simple Data Augmentation Method for
+    ASR](https://arxiv.org/abs/1904.08779). Note that this method is not optimized to run on TPU and should be run
     on CPU as part of the preprocessing during training.
 
     Args:
@@ -866,8 +866,8 @@ class WavLMEncoderStableLayerNorm(nn.Module):
 
 class WavLMGumbelVectorQuantizer(nn.Module):
     """
-    Vector quantization using gumbel softmax. See `CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX
-    <https://arxiv.org/pdf/1611.01144.pdf>`__ for more information.
+    Vector quantization using gumbel softmax. See [CATEGORICAL REPARAMETERIZATION WITH
+    GUMBEL-SOFTMAX](https://arxiv.org/pdf/1611.01144.pdf) for more information.
     """
 
     def __init__(self, config):
@@ -1167,8 +1167,8 @@ class WavLMModel(WavLMPreTrainedModel):
         attention_mask: Optional[torch.LongTensor] = None,
     ):
         """
-        Masks extracted features along time axis and/or along feature axis according to `SpecAugment
-        <https://arxiv.org/abs/1904.08779>`__ .
+        Masks extracted features along time axis and/or along feature axis according to
+        [SpecAugment](https://arxiv.org/abs/1904.08779).
         """
 
         # `config.apply_spec_augment` can set masking to False

templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/{{cookiecutter.lowercase_modelname}}.mdx

+<!--Copyright 2021 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+-->
+
+# {{cookiecutter.modelname}}
+
+## Overview
+
+The {{cookiecutter.modelname}} model was proposed in [<INSERT PAPER NAME HERE>(<INSERT PAPER LINK HERE>)  by <INSERT AUTHORS HERE>. <INSERT SHORT SUMMARY HERE>
+
+The abstract from the paper is the following:
+
+*<INSERT PAPER ABSTRACT HERE>*
+
+Tips:
+
+<INSERT TIPS ABOUT MODEL HERE>
+
+This model was contributed by [INSERT YOUR HF USERNAME HERE](<https://huggingface.co/<INSERT YOUR HF USERNAME HERE>)`. The original code can be found [here](<INSERT LINK TO GITHUB REPO HERE>).
+
+## XLNetConfig
+
+[[autodoc]] XLNetConfig
+
+## {{cookiecutter.camelcase_modelname}}Config
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}Config
+
+
+## {{cookiecutter.camelcase_modelname}}Tokenizer
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}Tokenizer
+    - build_inputs_with_special_tokens
+    - get_special_tokens_mask
+    - create_token_type_ids_from_sequences
+    - save_vocabulary
+
+
+## {{cookiecutter.camelcase_modelname}}TokenizerFast
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}TokenizerFast
+
+
+{% if "PyTorch" in cookiecutter.generate_tensorflow_pytorch_and_flax -%}
+## {{cookiecutter.camelcase_modelname}}Model
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}Model
+    - forward
+
+{% if cookiecutter.is_encoder_decoder_model == "False" %}
+## {{cookiecutter.camelcase_modelname}}ForCausalLM
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForCausalLM
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForMaskedLM
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForMaskedLM
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForSequenceClassification
+
+[[autodoc]] transformers.{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+    - forward
+
+## {{cookiecutter.camelcase_modelname}}ForMultipleChoice
+
+[[autodoc]] transformers.{{cookiecutter.camelcase_modelname}}ForMultipleChoice
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForTokenClassification
+
+[[autodoc]] transformers.{{cookiecutter.camelcase_modelname}}ForTokenClassification
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+    - forward
+
+{%- else %}
+## {{cookiecutter.camelcase_modelname}}ForConditionalGeneration
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForConditionalGeneration
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForSequenceClassification
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForSequenceClassification
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+    - forward
+
+
+## {{cookiecutter.camelcase_modelname}}ForCausalLM
+
+[[autodoc]] {{cookiecutter.camelcase_modelname}}ForCausalLM
+    - forward
+
+
+{% endif -%}
+{% endif -%}
+{% if "TensorFlow" in cookiecutter.generate_tensorflow_pytorch_and_flax -%}
+
+## TF{{cookiecutter.camelcase_modelname}}Model
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}Model
+    - call
+
+{% if cookiecutter.is_encoder_decoder_model == "False" %}
+## TF{{cookiecutter.camelcase_modelname}}ForMaskedLM
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForMaskedLM
+    - call
+
+
+## TF{{cookiecutter.camelcase_modelname}}ForCausalLM[[autodoc]] 
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForCausalLM
+    - call
+
+
+## TF{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+    - call
+
+
+## TF{{cookiecutter.camelcase_modelname}}ForMultipleChoice
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForMultipleChoice
+    - call
+
+
+## TF{{cookiecutter.camelcase_modelname}}ForTokenClassification
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForTokenClassification
+    - call
+
+
+## TF{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+    - call
+
+
+{%- else %}
+## TF{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
+
+[[autodoc]] TF{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
+    - call
+
+
+{% endif -%}
+{% endif -%}
+
+{% if "Flax" in cookiecutter.generate_tensorflow_pytorch_and_flax -%}
+
+## Flax{{cookiecutter.camelcase_modelname}}Model
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}Model
+    - call
+
+{% if cookiecutter.is_encoder_decoder_model == "False" %}
+## Flax{{cookiecutter.camelcase_modelname}}ForMaskedLM
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForMaskedLM
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForCausalLM
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForCausalLM
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForMultipleChoice
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForMultipleChoice
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForTokenClassification
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForTokenClassification
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+    - call
+
+
+{%- else %}
+## Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
+    - call
+
+
+## Flax{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
+
+[[autodoc]] Flax{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
+    - call
+
+
+{% endif -%}
+{% endif -%}

templates/adding_a_new_model/cookiecutter-template-{{cookiecutter.modelname}}/{{cookiecutter.lowercase_modelname}}.rst

-.. 
-    Copyright 2020 The HuggingFace Team. All rights reserved.
-
-    Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
-    the License. You may obtain a copy of the License at
-
-        http://www.apache.org/licenses/LICENSE-2.0
-
-    Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
-    an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
-    specific language governing permissions and limitations under the License.
-
-{{cookiecutter.modelname}}
------------------------------------------------------------------------------------------------------------------------
-
-Overview
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-The {{cookiecutter.modelname}} model was proposed in `<INSERT PAPER NAME HERE>
-<<INSERT PAPER LINK HERE>>`__  by <INSERT AUTHORS HERE>. <INSERT SHORT SUMMARY HERE>
-
-The abstract from the paper is the following:
-
-*<INSERT PAPER ABSTRACT HERE>*
-
-Tips:
-
-<INSERT TIPS ABOUT MODEL HERE>
-
-This model was contributed by `<INSERT YOUR HF USERNAME HERE> 
-<https://huggingface.co/<INSERT YOUR HF USERNAME HERE>>`__. The original code can be found `here 
-<<INSERT LINK TO GITHUB REPO HERE>>`__.
-
-{{cookiecutter.camelcase_modelname}}Config
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}Config
-    :members:
-
-
-{{cookiecutter.camelcase_modelname}}Tokenizer
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}Tokenizer
-    :members: build_inputs_with_special_tokens, get_special_tokens_mask,
-        create_token_type_ids_from_sequences, save_vocabulary
-
-
-{{cookiecutter.camelcase_modelname}}TokenizerFast
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}TokenizerFast
-    :members:
-
-
-{% if "PyTorch" in cookiecutter.generate_tensorflow_pytorch_and_flax -%}
-{{cookiecutter.camelcase_modelname}}Model
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}Model
-    :members: forward
-
-{% if cookiecutter.is_encoder_decoder_model == "False" %}
-{{cookiecutter.camelcase_modelname}}ForCausalLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForCausalLM
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForMaskedLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForMaskedLM
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForMultipleChoice
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForMultipleChoice
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForTokenClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForTokenClassification
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-    :members: forward
-
-{%- else %}
-{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-    :members: forward
-
-
-{{cookiecutter.camelcase_modelname}}ForCausalLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.{{cookiecutter.camelcase_modelname}}ForCausalLM
-    :members: forward
-
-
-{% endif -%}
-{% endif -%}
-{% if "TensorFlow" in cookiecutter.generate_tensorflow_pytorch_and_flax -%}
-
-TF{{cookiecutter.camelcase_modelname}}Model
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}Model
-    :members: call
-
-{% if cookiecutter.is_encoder_decoder_model == "False" %}
-TF{{cookiecutter.camelcase_modelname}}ForMaskedLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForMaskedLM
-    :members: call
-
-
-TF{{cookiecutter.camelcase_modelname}}ForCausalLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForCausalLM
-    :members: call
-
-
-TF{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-    :members: call
-
-
-TF{{cookiecutter.camelcase_modelname}}ForMultipleChoice
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForMultipleChoice
-    :members: call
-
-
-TF{{cookiecutter.camelcase_modelname}}ForTokenClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForTokenClassification
-    :members: call
-
-
-TF{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-    :members: call
-
-
-{%- else %}
-TF{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.TF{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
-    :members: call
-
-
-{% endif -%}
-{% endif -%}
-
-{% if "Flax" in cookiecutter.generate_tensorflow_pytorch_and_flax -%}
-
-Flax{{cookiecutter.camelcase_modelname}}Model
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}Model
-    :members: call
-
-{% if cookiecutter.is_encoder_decoder_model == "False" %}
-Flax{{cookiecutter.camelcase_modelname}}ForMaskedLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForMaskedLM
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForCausalLM
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForCausalLM
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForMultipleChoice
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForMultipleChoice
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForTokenClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForTokenClassification
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-    :members: call
-
-
-{%- else %}
-Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForSequenceClassification
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForQuestionAnswering
-    :members: call
-
-
-Flax{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-.. autoclass:: transformers.Flax{{cookiecutter.camelcase_modelname}}ForConditionalGeneration
-    :members: call
-
-
-{% endif -%}
-{% endif -%}

utils/check_table.py

@@ -208,15 +208,15 @@ def get_onnx_model_list():
 def check_onnx_model_list(overwrite=False):
     """Check the model list in the serialization.rst is consistent with the state of the lib and maybe `overwrite`."""
     current_list, start_index, end_index, lines = _find_text_in_file(
-        filename=os.path.join(PATH_TO_DOCS, "serialization.rst"),
-        start_prompt="    This table is automatically generated by make style, do not fill manually!",
+        filename=os.path.join(PATH_TO_DOCS, "serialization.mdx"),
+        start_prompt="<!--This table is automatically generated by make style, do not fill manually!-->",
         end_prompt="This conversion is handled with the PyTorch version of models ",
     )
     new_list = get_onnx_model_list()
 
     if current_list != new_list:
         if overwrite:
-            with open(os.path.join(PATH_TO_DOCS, "serialization.rst"), "w", encoding="utf-8", newline="\n") as f:
+            with open(os.path.join(PATH_TO_DOCS, "serialization.mdx"), "w", encoding="utf-8", newline="\n") as f:
                 f.writelines(lines[:start_index] + [new_list] + lines[end_index:])
         else:
             raise ValueError("The list of ONNX-supported models needs an update. Run `make fix-copies` to fix this.")