# modeling_mixture_of_recursions.py
# Create this file in your repository root

import torch
from transformers import PreTrainedModel
from transformers.modeling_outputs import CausalLMOutputWithPast
from typing import Optional, Tuple

# Import your existing model
try:
    from model_slm import *  # Import everything from your existing model file
except:
    pass  # Will work when uploaded to HF

from .configuration_mixture_of_recursions import MixtureOfRecursionsConfig


class MixtureOfRecursionsPreTrainedModel(PreTrainedModel):
    """
    An abstract class to handle weights initialization and a simple interface for downloading and loading pretrained models.
    """
    config_class = MixtureOfRecursionsConfig
    base_model_prefix = "model"
    supports_gradient_checkpointing = False
    _no_split_modules = []

    def _init_weights(self, module):
        """Initialize the weights"""
        if isinstance(module, torch.nn.Linear):
            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
            if module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, torch.nn.Embedding):
            module.weight.data.normal_(mean=0.0, std=self.config.initializer_range)
            if module.padding_idx is not None:
                module.weight.data[module.padding_idx].zero_()
        elif isinstance(module, torch.nn.LayerNorm):
            module.bias.data.zero_()
            module.weight.data.fill_(1.0)


class MixtureOfRecursionsModel(MixtureOfRecursionsPreTrainedModel):
    """
    Wrapper around your existing model to make it compatible with Transformers
    """
    def __init__(self, config):
        super().__init__(config)
        self.config = config
        
        # This should match your actual model initialization from model_slm.py
        # Replace this with your actual model class name
        # For example: self.model = YourModelClass(config)
        
        # Placeholder - update with your actual model architecture
        self.vocab_size = config.vocab_size
        self.hidden_size = config.hidden_size
        
        self.post_init()

    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        attention_mask: Optional[torch.FloatTensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        labels: Optional[torch.LongTensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ):
        # Your forward pass logic from model_slm.py
        # This is a placeholder - replace with your actual forward implementation
        pass


class MixtureOfRecursionsForCausalLM(MixtureOfRecursionsPreTrainedModel):
    """
    Causal LM head wrapper for your model
    """
    _tied_weights_keys = ["lm_head.weight"]

    def __init__(self, config):
        super().__init__(config)
        self.model = MixtureOfRecursionsModel(config)
        self.vocab_size = config.vocab_size
        self.lm_head = torch.nn.Linear(config.hidden_size, config.vocab_size, bias=False)
        
        self.post_init()

    def get_input_embeddings(self):
        return self.model.embed_tokens if hasattr(self.model, 'embed_tokens') else None

    def set_input_embeddings(self, value):
        if hasattr(self.model, 'embed_tokens'):
            self.model.embed_tokens = value

    def get_output_embeddings(self):
        return self.lm_head

    def set_output_embeddings(self, new_embeddings):
        self.lm_head = new_embeddings

    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        attention_mask: Optional[torch.FloatTensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        labels: Optional[torch.LongTensor] = None,
        use_cache: Optional[bool] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ):
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        # Forward pass through model
        outputs = self.model(
            input_ids=input_ids,
            attention_mask=attention_mask,
            position_ids=position_ids,
            past_key_values=past_key_values,
            inputs_embeds=inputs_embeds,
            use_cache=use_cache,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        hidden_states = outputs[0] if isinstance(outputs, tuple) else outputs.last_hidden_state
        logits = self.lm_head(hidden_states)

        loss = None
        if labels is not None:
            # Shift for causal language modeling
            shift_logits = logits[..., :-1, :].contiguous()
            shift_labels = labels[..., 1:].contiguous()
            loss_fct = torch.nn.CrossEntropyLoss()
            loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))

        if not return_dict:
            output = (logits,) + outputs[1:]
            return ((loss,) + output) if loss is not None else output

        return CausalLMOutputWithPast(
            loss=loss,
            logits=logits,
            past_key_values=outputs.past_key_values if hasattr(outputs, 'past_key_values') else None,
            hidden_states=outputs.hidden_states if hasattr(outputs, 'hidden_states') else None,
            attentions=outputs.attentions if hasattr(outputs, 'attentions') else None,
        )

    def prepare_inputs_for_generation(
        self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs
    ):
        if past_key_values:
            input_ids = input_ids[:, -1:]

        position_ids = kwargs.get("position_ids", None)
        if attention_mask is not None and position_ids is None:
            position_ids = attention_mask.long().cumsum(-1) - 1
            position_ids.masked_fill_(attention_mask == 0, 1)
            if past_key_values:
                position_ids = position_ids[:, -1].unsqueeze(-1)

        if inputs_embeds is not None and past_key_values is None:
            model_inputs = {"inputs_embeds": inputs_embeds}
        else:
            model_inputs = {"input_ids": input_ids}

        model_inputs.update(
            {
                "position_ids": position_ids,
                "past_key_values": past_key_values,
                "use_cache": kwargs.get("use_cache"),
                "attention_mask": attention_mask,
            }
        )
        return model_inputs