fix more doctests (#22292)

* fix more doctests

* fix style

---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

src/transformers/models/mvp/tokenization_mvp_fast.py

@@ -58,12 +58,14 @@ class MvpTokenizerFast(PreTrainedTokenizerFast):
     This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will
     be encoded differently whether it is at the beginning of the sentence (without space) or not:
 
-    ```
+    ```python
     >>> from transformers import MvpTokenizerFast
+
     >>> tokenizer = MvpTokenizerFast.from_pretrained("RUCAIBox/mvp")
-    >>> tokenizer("Hello world")['input_ids']
+    >>> tokenizer("Hello world")["input_ids"]
     [0, 31414, 232, 2]
-    >>> tokenizer(" Hello world")['input_ids']
+
+    >>> tokenizer(" Hello world")["input_ids"]
     [0, 20920, 232, 2]
     ```
 

src/transformers/models/roberta/tokenization_roberta.py

@@ -111,12 +111,14 @@ class RobertaTokenizer(PreTrainedTokenizer):
     This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will
     be encoded differently whether it is at the beginning of the sentence (without space) or not:
 
-    ```
+    ```python
     >>> from transformers import RobertaTokenizer
+
     >>> tokenizer = RobertaTokenizer.from_pretrained("roberta-base")
-    >>> tokenizer("Hello world")['input_ids']
+    >>> tokenizer("Hello world")["input_ids"]
     [0, 31414, 232, 2]
-    >>> tokenizer(" Hello world")['input_ids']
+
+    >>> tokenizer(" Hello world")["input_ids"]
     [0, 20920, 232, 2]
     ```
 

src/transformers/models/roberta/tokenization_roberta_fast.py

@@ -81,12 +81,14 @@ class RobertaTokenizerFast(PreTrainedTokenizerFast):
     This tokenizer has been trained to treat spaces like parts of the tokens (a bit like sentencepiece) so a word will
     be encoded differently whether it is at the beginning of the sentence (without space) or not:
 
-    ```
+    ```python
     >>> from transformers import RobertaTokenizerFast
+
     >>> tokenizer = RobertaTokenizerFast.from_pretrained("roberta-base")
-    >>> tokenizer("Hello world")['input_ids']
+    >>> tokenizer("Hello world")["input_ids"]
     [0, 31414, 232, 2]
-    >>> tokenizer(" Hello world")['input_ids']
+
+    >>> tokenizer(" Hello world")["input_ids"]
     [0, 20920, 232, 2]
     ```
 

utils/documentation_tests.txt

@@ -444,3 +444,26 @@ src/transformers/models/wav2vec2/processing_wav2vec2.py
 src/transformers/models/wav2vec2_with_lm/processing_wav2vec2_with_lm.py
 src/transformers/models/whisper/processing_whisper.py
 src/transformers/models/x_clip/processing_x_clip.py
+src/transformers/models/bart/tokenization_bart.py
+src/transformers/models/bart/tokenization_bart_fast.py
+src/transformers/models/blenderbot/tokenization_blenderbot.py
+src/transformers/models/blenderbot/tokenization_blenderbot_fast.py
+src/transformers/models/bloom/tokenization_bloom_fast.py
+src/transformers/models/codegen/tokenization_codegen.py
+src/transformers/models/codegen/tokenization_codegen_fast.py
+src/transformers/models/deberta/tokenization_deberta.py
+src/transformers/models/deberta/tokenization_deberta_fast.py
+src/transformers/models/gpt2/tokenization_gpt2.py
+src/transformers/models/gpt2/tokenization_gpt2_fast.py
+src/transformers/models/gpt_neox/tokenization_gpt_neox_fast.py
+src/transformers/models/gpt_sw3/tokenization_gpt_sw3.py
+src/transformers/models/jukebox/tokenization_jukebox.py
+src/transformers/models/led/tokenization_led.py
+src/transformers/models/led/tokenization_led_fast.py
+src/transformers/models/longformer/tokenization_longformer.py
+src/transformers/models/longformer/tokenization_longformer_fast.py
+src/transformers/models/luke/tokenization_luke.py
+src/transformers/models/mvp/tokenization_mvp.py
+src/transformers/models/mvp/tokenization_mvp_fast.py
+src/transformers/models/roberta/tokenization_roberta.py
+src/transformers/models/roberta/tokenization_roberta_fast.py