Remove more unused attributes in config classes (#21327)

* remove unused classifier_dropout

* remove unused dropout

* remove unused pooler_fn

* remove unnecessary is_encoder_decoder

* remove unnecessary drop_rate

* remove unused classifier_dropout

* remove unused classifier_dropout

* remove unused dropout

* remove unused dropout

* remove unused summary_* attributes

* remove unused tie_word_embeddings

* remove unused summary_* attributes

* fix

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

src/transformers/models/trocr/configuration_trocr.py

@@ -63,8 +63,6 @@ class TrOCRConfig(PretrainedConfig):
             The dropout ratio for the attention probabilities.
         activation_dropout (`float`, *optional*, defaults to 0.0):
             The dropout ratio for activations inside the fully connected layer.
-        classifier_dropout (`float`, *optional*, defaults to 0.0):
-            The dropout ratio for classifier.
         init_std (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         decoder_layerdrop (`float`, *optional*, defaults to 0.0):
@@ -114,7 +112,6 @@ class TrOCRConfig(PretrainedConfig):
         attention_dropout=0.0,
         activation_dropout=0.0,
         decoder_start_token_id=2,
-        classifier_dropout=0.0,
         init_std=0.02,
         decoder_layerdrop=0.0,
         use_cache=True,
@@ -136,7 +133,6 @@ class TrOCRConfig(PretrainedConfig):
         self.dropout = dropout
         self.attention_dropout = attention_dropout
         self.activation_dropout = activation_dropout
-        self.classifier_dropout = classifier_dropout
         self.init_std = init_std
         self.decoder_layerdrop = decoder_layerdrop
         self.use_cache = use_cache

src/transformers/models/whisper/configuration_whisper.py

@@ -129,8 +129,6 @@ class WhisperConfig(PretrainedConfig):
             Begin of stream token id.
         eos_token_id (`int`, *optional*, defaults to 50257):
             End of stream token id.
-        tie_word_embeddings (`bool`, *optional*, defaults to `True`):
-            Whether to tie input and output embeddings.
         suppress_tokens (`List[int]`, *optional*):
             A list containing the non-speech tokens that will be used by the logit processor in the `generate`
             function. NON_SPEECH_TOKENS and NON_SPEECH_TOKENS_MULTI each correspond to the `english-only` and the
@@ -185,7 +183,6 @@ class WhisperConfig(PretrainedConfig):
         pad_token_id=50256,
         bos_token_id=50257,
         eos_token_id=50256,
-        tie_word_embeddings=True,
         suppress_tokens=None,
         begin_suppress_tokens=[220, 50256],
         **kwargs
@@ -209,7 +206,6 @@ class WhisperConfig(PretrainedConfig):
         self.use_cache = use_cache
         self.num_hidden_layers = encoder_layers
         self.scale_embedding = scale_embedding  # scale factor will be sqrt(d_model) if True
-        self.tie_word_embeddings = tie_word_embeddings
         self.max_source_positions = max_source_positions
         self.max_target_positions = max_target_positions
         super().__init__(
@@ -218,7 +214,6 @@ class WhisperConfig(PretrainedConfig):
             eos_token_id=eos_token_id,
             is_encoder_decoder=is_encoder_decoder,
             decoder_start_token_id=decoder_start_token_id,
-            tie_word_embeddings=tie_word_embeddings,
             suppress_tokens=suppress_tokens,
             begin_suppress_tokens=begin_suppress_tokens,
             **kwargs,

src/transformers/models/x_clip/configuration_x_clip.py

@@ -62,8 +62,6 @@ class XCLIPTextConfig(PretrainedConfig):
             The epsilon used by the layer normalization layers.
         attention_dropout (`float`, *optional*, defaults to 0.0):
             The dropout ratio for the attention probabilities.
-        dropout (`float`, *optional*, defaults to 0.0):
-            The dropout probabilitiy for all fully connected layers in the embeddings, encoder, and pooler.
         initializer_range (`float`, *optional*, defaults to 0.02):
             The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
         initializer_factor (`float``, *optional*, defaults to 1):
@@ -96,7 +94,6 @@ class XCLIPTextConfig(PretrainedConfig):
         max_position_embeddings=77,
         hidden_act="quick_gelu",
         layer_norm_eps=1e-5,
-        dropout=0.0,
         attention_dropout=0.0,
         initializer_range=0.02,
         initializer_factor=1.0,
@@ -110,7 +107,6 @@ class XCLIPTextConfig(PretrainedConfig):
         self.vocab_size = vocab_size
         self.hidden_size = hidden_size
         self.intermediate_size = intermediate_size
-        self.dropout = dropout
         self.num_hidden_layers = num_hidden_layers
         self.num_attention_heads = num_attention_heads
         self.max_position_embeddings = max_position_embeddings
@@ -176,8 +172,6 @@ class XCLIPVisionConfig(PretrainedConfig):
             `"relu"`, `"selu"`, `"gelu_new"` and ``"quick_gelu"` are supported.
         layer_norm_eps (`float`, *optional*, defaults to 1e-5):
             The epsilon used by the layer normalization layers.
-        dropout (`float`, *optional*, defaults to 0.0):
-            The dropout probabilitiy for all fully connected layers in the embeddings, encoder, and pooler.
         attention_dropout (`float`, *optional*, defaults to 0.0):
             The dropout ratio for the attention probabilities.
         initializer_range (`float`, *optional*, defaults to 0.02):
@@ -221,7 +215,6 @@ class XCLIPVisionConfig(PretrainedConfig):
         num_frames=8,
         hidden_act="quick_gelu",
         layer_norm_eps=1e-5,
-        dropout=0.0,
         attention_dropout=0.0,
         initializer_range=0.02,
         initializer_factor=1.0,
@@ -232,7 +225,6 @@ class XCLIPVisionConfig(PretrainedConfig):
 
         self.hidden_size = hidden_size
         self.intermediate_size = intermediate_size
-        self.dropout = dropout
         self.num_hidden_layers = num_hidden_layers
         self.num_attention_heads = num_attention_heads
         self.mit_hidden_size = mit_hidden_size