transformers 5.0.0rc1__py3-none-any.whl → 5.0.0rc2__py3-none-any.whl

transformers/models/ernie4_5_vl_moe/configuration_ernie4_5_vl_moe.py

@@ -0,0 +1,330 @@
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+#           This file was automatically generated from src/transformers/models/ernie4_5_vl_moe/modular_ernie4_5_vl_moe.py.
+#               Do NOT edit this file manually as any edits will be overwritten by the generation of
+#             the file from the modular. If any change should be done, please apply the change to the
+#                          modular_ernie4_5_vl_moe.py file directly. One of our CI enforces this.
+#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+# coding=utf-8
+# Copyright 2025 Baidu and HuggingFace Inc. 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.
+from ...configuration_utils import PreTrainedConfig, layer_type_validation
+
+
+class Ernie4_5_VL_MoeVisionConfig(PreTrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of the [`Ernie4_5_VL_MoeVisionTransformerPretrainedModel`].
+    It is used to instantiate the vision models portion of the complete Ernie4.5-VL Moe model according to the specified
+    arguments, defining the model architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+    Args:
+        depth (`int`, *optional*, defaults to 32):
+            Number of layers (depth) in the model.
+        hidden_size (`int`, *optional*, defaults to 1280):
+            Dimensionality of the encoder layers and the pooler layer.
+        hidden_act (`str` or `function`, *optional*, defaults to `"quick_gelu"`):
+            The non-linear activation function (function or string) in the encoder and pooler.
+        intermediate_size (`int`, *optional*, defaults to 5120):
+            Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
+        num_heads (`int`, *optional*, defaults to 16):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        in_channels (`int`, *optional*, defaults to 3):
+            The number of input channels.
+        patch_size (`int`, *optional*, defaults to 14):
+            The size (resolution) of each patch.
+        spatial_merge_size (`int`, *optional*, defaults to 2):
+            The size used for merging spatial dimensions.
+        temporal_merge_size (`int`, *optional*, defaults to 2):
+            The size used for merge along the temporal dimension.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
+            The epsilon used by the rms normalization layers.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+    """
+
+    model_type = "ernie4_5_vl_moe_vision"
+    base_config_key = "vision_config"
+
+    base_model_tp_plan = {
+        "blocks.*.attn.qkv": "colwise",
+        "blocks.*.attn.proj": "rowwise",
+        "blocks.*.mlp.fc1": "colwise",
+        "blocks.*.mlp.fc2": "rowwise",
+    }
+
+    def __init__(
+        self,
+        depth=32,
+        hidden_size=1280,
+        hidden_act="quick_gelu",
+        intermediate_size=4 * 1280,
+        num_heads=16,
+        in_channels=3,
+        patch_size=14,
+        spatial_merge_size=2,
+        temporal_merge_size=2,
+        rms_norm_eps=1e-6,
+        initializer_range=0.02,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.depth = depth
+        self.hidden_size = hidden_size
+        self.hidden_act = hidden_act
+        self.num_heads = num_heads
+        self.in_channels = in_channels
+        self.patch_size = patch_size
+        self.spatial_merge_size = spatial_merge_size
+        self.initializer_range = initializer_range
+
+        self.intermediate_size = intermediate_size
+        self.temporal_merge_size = temporal_merge_size
+        self.rms_norm_eps = rms_norm_eps
+
+
+class Ernie4_5_VL_MoeTextConfig(PreTrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`Ernie4_5_VL_MoeTextModel`]. It is used to instantiate a
+    the text model portion of the complete Ernie4.5-VL Moe model according to the specified arguments, defining the model architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 103424):
+            Vocabulary size of the Ernie 4.5 VL model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`Ernie4_5_VL_MoeTextModel`]
+        hidden_size (`int`, *optional*, defaults to 2560):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 12288):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 28):
+            Number of hidden layers in the Transformer encoder.
+        num_attention_heads (`int`, *optional*, defaults to 20):
+            Number of attention heads for each attention layer in the Transformer encoder.
+        num_key_value_heads (`int`, *optional*, defaults to 4):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details, check out [this
+            paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `4`.
+        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
+            The non-linear activation function (function or string) in the decoder.
+        max_position_embeddings (`int`, *optional*, defaults to 131072):
+            The maximum sequence length that this model might ever be used with.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-05):
+            The epsilon used by the rms normalization layers.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        use_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use a bias in any of the projections including mlp and attention for example.
+        tie_word_embeddings (`bool`, *optional*, defaults to `True`):
+            Whether the model's input and output word embeddings should be tied.
+        rope_parameters (`RopeParameters`, *optional*):
+            Dictionary containing the configuration parameters for the RoPE embeddings. The dictionaty should contain
+            a value for `rope_theta` and optionally parameters used for scaling in case you want to use RoPE
+            with longer `max_position_embeddings`.
+        mlp_layer_types (`list`, *optional*):
+            MLP (Moe vs Dense) pattern for each layer.
+        moe_intermediate_size (`list[int]`, *optional*, defaults to `[1536, 512]`):
+            Intermediate size of the routed experts; differs between text (first) and image (second) experts.
+        moe_k (`int`, *optional*, defaults to 6):
+            Number of selected experts.
+        moe_num_experts (`int`, *optional*, defaults to 64):
+            Number of routed experts.
+        moe_num_shared_experts (`int`, *optional*, defaults to 2):
+            The number of experts that are shared for all MoE forwards.
+        moe_norm_min (`float`, *optional*, defaults to 1e-12):
+            Minimum division value during routing normalization.
+        output_router_logits (`bool`, *optional*, defaults to `False`):
+            Whether or not the router logits should be returned by the model. Enabling this will also
+            allow the model to output the auxiliary loss, including load balancing loss and router z-loss.
+        router_aux_loss_coef (`float`, *optional*, defaults to 0.001):
+            The aux loss factor for the total loss.
+    """
+
+    model_type = "ernie4_5_vl_moe_text"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    attribute_map = {"num_experts": "moe_num_experts", "num_experts_per_tok": "moe_k"}
+    default_theta = 500000.0
+
+    base_model_tp_plan = {
+        "layers.*.self_attn.q_proj": "colwise",
+        "layers.*.self_attn.k_proj": "colwise",
+        "layers.*.self_attn.v_proj": "colwise",
+        "layers.*.self_attn.o_proj": "rowwise",
+        "layers.*.mlp.shared_experts.gate_proj": "colwise",
+        "layers.*.mlp.shared_experts.up_proj": "colwise",
+        "layers.*.mlp.shared_experts.down_proj": "rowwise",
+        "layers.*.mlp.gate_proj": "colwise",
+        "layers.*.mlp.up_proj": "colwise",
+        "layers.*.mlp.down_proj": "rowwise",
+    }
+    base_model_pp_plan = {
+        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
+        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
+        "norm": (["hidden_states"], ["hidden_states"]),
+    }
+    base_config_key = "text_config"
+
+    def __init__(
+        self,
+        vocab_size=103424,
+        hidden_size=2560,
+        intermediate_size=12288,
+        num_hidden_layers=28,
+        num_attention_heads=20,
+        num_key_value_heads=4,
+        hidden_act="silu",
+        max_position_embeddings=131072,
+        initializer_range=0.02,
+        rms_norm_eps=1e-5,
+        use_cache=True,
+        use_bias=False,
+        tie_word_embeddings=True,
+        rope_parameters=None,
+        mlp_layer_types=None,
+        moe_intermediate_size=None,
+        moe_k=6,
+        moe_num_experts=64,
+        moe_num_shared_experts=2,
+        moe_norm_min=1e-12,
+        output_router_logits=False,
+        router_aux_loss_coef=0.001,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.max_position_embeddings = max_position_embeddings
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.use_bias = use_bias
+        self.rope_parameters = rope_parameters
+
+        # Default to MoE from the second layer and on
+        self.mlp_layer_types = mlp_layer_types
+        if self.mlp_layer_types is None:
+            self.mlp_layer_types = ["dense"] + ["sparse"] * (self.num_hidden_layers - 1)
+        layer_type_validation(self.mlp_layer_types, self.num_hidden_layers, attention=False)
+
+        self.moe_intermediate_size = moe_intermediate_size
+        if self.moe_intermediate_size is None:
+            self.moe_intermediate_size = [1536, 512]
+        self.moe_k = moe_k
+        self.moe_num_experts = moe_num_experts
+        self.moe_num_shared_experts = moe_num_shared_experts
+        self.moe_norm_min = moe_norm_min
+        self.output_router_logits = output_router_logits
+        self.router_aux_loss_coef = router_aux_loss_coef
+
+        super().__init__(
+            tie_word_embeddings=tie_word_embeddings, ignore_keys_at_rope_validation={"mrope_section"}, **kwargs
+        )
+
+
+class Ernie4_5_VL_MoeConfig(PreTrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`Ernie4_5_VL_MoeModel`]. It is used to instantiate a
+    Ernie4.5-VL MoE model according to the specified arguments, defining the model architecture. Instantiating a configuration
+    with the defaults will yield a similar configuration to that of
+    Ernie 4.5 VL 28B A3B [baidu/ERNIE-4.5-VL-28B-A3B-PT](https://huggingface.co/baidu/ERNIE-4.5-VL-28B-A3B-PT).
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+    Args:
+        text_config (`Union[PreTrainedConfig, dict]`, *optional*, defaults to `Ernie4_5_VL_MoeTextConfig`):
+            The config object or dictionary of the text backbone.
+        vision_config (`Union[PreTrainedConfig, dict]`,  *optional*, defaults to `Ernie4_5_VL_MoeVisionConfig`):
+            The config object or dictionary of the vision backbone.
+        image_start_token_id (`int`, *optional*, defaults to 101304):
+            The image token index to encode the start of image.
+        image_end_token_id (`int`, *optional*, defaults to 101305):
+            The image token index to encode the end of image.
+        image_token_id (`int`, *optional*, defaults to 100295):
+            The image token index to encode the image prompt.
+        video_start_token_id (`int`, *optional*, defaults to 101306):
+            The video token index to encode the start of video.
+        video_end_token_id (`int`, *optional*, defaults to 101307):
+            The video token index to encode the end of video.
+        video_token_id (`int`, *optional*, defaults to 103367):
+            The video token index to encode the video prompt.
+
+    ```python
+    >>> from transformers import Ernie4_5_VL_MoeForConditionalGeneration, Ernie4_5_VL_MoeConfig
+
+    >>> # Initializing a Ernie4_5_VL_Moe style configuration
+    >>> configuration = Ernie4_5_VL_MoeConfig()
+
+    >>> # Initializing a model from the Ernie 4.5 VL 28B A3B configuration
+    >>> model = Ernie4_5_VL_MoeForConditionalGeneration(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "ernie4_5_vl_moe"
+    sub_configs = {"vision_config": Ernie4_5_VL_MoeVisionConfig, "text_config": Ernie4_5_VL_MoeTextConfig}
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        text_config=None,
+        vision_config=None,
+        image_start_token_id=101304,
+        image_end_token_id=101305,
+        image_token_id=100295,
+        video_start_token_id=101306,
+        video_end_token_id=101307,
+        video_token_id=103367,
+        **kwargs,
+    ):
+        if isinstance(vision_config, dict):
+            self.vision_config = self.sub_configs["vision_config"](**vision_config)
+        elif isinstance(vision_config, Ernie4_5_VL_MoeVisionConfig):
+            self.vision_config = vision_config
+        elif vision_config is None:
+            self.vision_config = self.sub_configs["vision_config"]()
+
+        if isinstance(text_config, dict):
+            self.text_config = self.sub_configs["text_config"](**text_config)
+        elif isinstance(text_config, Ernie4_5_VL_MoeTextConfig):
+            self.text_config = text_config
+        elif text_config is None:
+            self.text_config = self.sub_configs["text_config"](**kwargs)
+
+        self.image_start_token_id = image_start_token_id
+        self.image_end_token_id = image_end_token_id
+        self.image_token_id = image_token_id
+        self.video_start_token_id = video_start_token_id
+        self.video_end_token_id = video_end_token_id
+        self.video_token_id = video_token_id
+
+        super().__init__(**kwargs)
+
+
+__all__ = ["Ernie4_5_VL_MoeConfig", "Ernie4_5_VL_MoeTextConfig", "Ernie4_5_VL_MoeVisionConfig"]