import os
from typing import Union
from transformers.configuration_utils import PretrainedConfig
from transformers.utils import logging
logger = logging.get_logger(__name__)


class Siglip2VisionConfig(PretrainedConfig):
    model_type = "siglip2_navit_rope_model"
    base_config_key = "vision_config"

    def __init__(
        self,
        hidden_size=768,
        intermediate_size=3072,
        num_hidden_layers=12,
        num_attention_heads=12,
        num_channels=3,
        patch_size=16,
        hidden_act="gelu_pytorch_tanh",
        layer_norm_eps=1e-6,
        preserve_original_pe=True,
        disable_rope=False,
        num_patches=1369,
        rope_theta=10000.0,
        **kwargs,
    ):
        super().__init__(**kwargs)

        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_channels = num_channels
        self.patch_size = patch_size
        self.hidden_act = hidden_act
        self.layer_norm_eps = layer_norm_eps
        self.preserve_original_pe = preserve_original_pe
        self.disable_rope = disable_rope
        self.num_patches = num_patches
        self.rope_theta = rope_theta