teacher_code/blip3o_fast_inference.py

from typing import List, Optional
import torch
import torch.nn as nn
from PIL import Image
import torch.nn.functional as F
from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer
from transformers import Qwen2_5_VLConfig, Qwen2ForCausalLM, Qwen2Config, Qwen2Model
from blip3o.constants import UND_IMAGE_TOKEN_IDX, DEFAULT_IMAGE_TOKEN
from ..llava_arch import LlavaMetaModel, LlavaMetaForCausalLM

from diffusers.utils.torch_utils import randn_tensor
from diffusers.pipelines.pipeline_utils import numpy_to_pil
import numpy as np
from diffusers.schedulers import FlowMatchEulerDiscreteScheduler
from tqdm import tqdm


class blip3oFastConfig(Qwen2Config):
    model_type = "blip3o_fast_inference"


class blip3oFastModel(LlavaMetaModel, Qwen2Model):
    config_class = blip3oFastConfig

    def __init__(self, config: Qwen2_5_VLConfig):
        super(blip3oFastModel, self).__init__(config)


class blip3oFastForInferenceLM(Qwen2ForCausalLM, LlavaMetaForCausalLM):
    config_class = blip3oFastConfig

    def __init__(self, config):
        super(blip3oFastForInferenceLM, self).__init__(config)
        config.model_type = "blip3o_qwen_inference"

        self.model = blip3oFastModel(config)
        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
        # Initialize weights and apply final processing
        self.post_init()

    def get_model(self):
        return self.model

    def visual(self, pixel_values: torch.Tensor, grid_thw: Optional[torch.Tensor] = None) -> torch.Tensor:
        image_features = self.get_model().get_vision_tower()(pixel_values)
        image_features = self.get_model().mm_projector(image_features)
        return image_features
    
    @torch.no_grad()
    def generate_image(
        self,
        input_ids: Optional[torch.Tensor] = None,
        attention_mask: Optional[torch.Tensor] = None,
        pixel_values: Optional[torch.Tensor] = None,
        image_grid_thw: Optional[torch.Tensor] = None,
        max_var: Optional[float] = None,
    ):  
        N_QUERY = self.get_n_query()            
        print("N_QUERY: ", N_QUERY)
        text_embeds = self.get_model().embed_tokens(input_ids)
        latent_queries = self.get_model().latent_queries.repeat(text_embeds.shape[0], 1, 1)


        if pixel_values is not None:
            und_image_idx = (input_ids == UND_IMAGE_TOKEN_IDX)
            pixel_values = pixel_values.type(self.visual.dtype)
            und_image_embeds = self.visual(pixel_values, grid_thw=image_grid_thw)
            text_embeds[und_image_idx] = und_image_embeds.to(text_embeds.device)[:und_image_idx.sum(), :]


        text_embeds = torch.cat([text_embeds, latent_queries], dim=1)
        attention_mask = torch.cat([attention_mask, torch.ones_like(latent_queries[:, :, 0])], dim=1)
        outputs = self.model(
            inputs_embeds=text_embeds,
            attention_mask=attention_mask,
            output_hidden_states=True,
            return_dict=True,
        )
        hidden_states = outputs.hidden_states[-1][:,-N_QUERY:,:]
        img_hidden_states = hidden_states 
        output_img = self.sample_images(img_hidden_states)
        return output_img

    def sample_images(
        self,
        pred_latents,
        guidance_scale: float = 3.0,
        num_inference_steps: int = 30,
        num_images_per_prompt: int = 1,
        return_tensor=False,
        **kwargs,
    ):
        device = pred_latents.device
        dtype = pred_latents.dtype

       
        img_hidden_states_null = torch.zeros_like(pred_latents, device=device, dtype=dtype)
        pred_latents = torch.cat([img_hidden_states_null, pred_latents], 0)
        batch_size = pred_latents.shape[0]
        latent_size = self.get_n_query()
        latent_channels = self.get_model().dit.config.in_channels

        latents = randn_tensor(
            shape=(batch_size * num_images_per_prompt, latent_channels, latent_size, latent_size),
            generator=None,
            device=device,
            dtype=dtype,
        )

        # set step values
        if isinstance(self.model.noise_scheduler, FlowMatchEulerDiscreteScheduler):
            sigmas = np.linspace(1.0, 1 / num_inference_steps, num_inference_steps)
            self.model.noise_scheduler.set_timesteps(num_inference_steps, sigmas=sigmas)
        else:
            self.model.noise_scheduler.set_timesteps(num_inference_steps)

        for t in tqdm(self.model.noise_scheduler.timesteps, desc="Sampling images"):
            latent_model_input = torch.cat([latents] * 2)
            latent_model_input = latent_model_input.to(pred_latents.dtype)
            if hasattr(self.model.noise_scheduler.timesteps, "scale_model_input"):
                latent_model_input = self.model.noise_scheduler.scale_model_input(latent_model_input, t)
            #print("latent_model_input: ", latent_model_input.shape)
            noise_pred = self.model.dit(
                hidden_states=latent_model_input,
                encoder_hidden_states=self.model.diffusion_connector(pred_latents),
                timestep=t.unsqueeze(0).expand(latent_model_input.shape[0]).to(latents.device),
                encoder_attention_mask=None
            ).sample

            noise_pred_uncond, noise_pred= noise_pred.chunk(2)
            noise_pred = noise_pred_uncond + guidance_scale * (noise_pred - noise_pred_uncond)
            latents = self.model.noise_scheduler.step(noise_pred, t, latents).prev_sample

        samples = self.decode_latents(latents.to(self.model.vae.dtype) if self.model.vae is not None else latents, return_tensor=return_tensor) 
        return samples

    @torch.no_grad()
    def decode_latents(self, latents, normalize=True, return_tensor=False):
        if self.model.vae is not None:
            latents = latents / self.model.vae.config.scaling_factor
            if "shift_factor" in self.model.vae.config and self.model.vae.config.shift_factor is not None:
                latents = latents + self.model.vae.config.shift_factor
            samples = self.model.vae.decode(latents).sample
        else:
            samples = latents
        if normalize:
            samples = (samples / 2 + 0.5).clamp(0, 1)
        else:
            samples = samples.clamp(-1, 1)
        if return_tensor:
            return samples
        samples = samples.cpu().permute(0, 2, 3, 1).float().numpy()
        samples = numpy_to_pil(samples)
        return samples
    
    def prepare_and_encode_inputs(
        self,
        inputs: List[str | Image.Image],
        tokenizer: AutoTokenizer,
        do_classifier_free_guidance: bool = False,
    ): 
        print("="*20, "prepare_and_encode_inputs", "="*20)
        # pdb.set_trace()
        device = self.get_model().device
        dtype = self.get_model().dtype

        has_image, has_text = False, False
        text_prompt, image_prompt = "", []
        img_processor = self.get_vision_tower().image_processor
        negative_prompt = {}

        for x in inputs:
            if isinstance(x, str):
                has_text = True
                text_prompt += x
            else:
                has_image = True
                text_prompt += DEFAULT_IMAGE_TOKEN
                image_prompt.append(img_processor.preprocess(x, return_tensors='pt')['pixel_values'])
        # pdb.set_trace()
        if len(image_prompt) == 0:
            image_prompt = None
        else:
            image_prompt = torch.cat(image_prompt)
            image_prompt = image_prompt.type(dtype).to(device)

        if has_image and not has_text:
            prompt = self.encode_images(image_prompt)
            # pdb.set_trace()
            if do_classifier_free_guidance:
                key = "[NULL_IMAGE]"
                if key not in negative_prompt:
                    negative_image = torch.zeros_like(image_prompt)
                    negative_prompt[key] = self.encode_images(negative_image)
                prompt = torch.cat([prompt, negative_prompt[key]], dim=0)
        else:
            prompt = self.generate_image(text=[text_prompt], image=image_prompt, tokenizer=tokenizer)
            if do_classifier_free_guidance:
                key = ""
                if key not in negative_prompt:
                    negative_prompt[key] = self.generate_image(text=[""], tokenizer=tokenizer)
                prompt = torch.cat([prompt, negative_prompt[key]], dim=0)
        
        gen_pooling = self.get_gen_pooling()
        n_query = self.get_n_query()
        num_img, _, c = prompt.shape
        if 'pool2d' in gen_pooling and has_text and not 'early' in gen_pooling:
            stride = int(gen_pooling.split('_')[1])
            sqrt_n = int(n_query**0.5)
            prompt = prompt.permute(0, 2, 1).reshape(num_img, -1, sqrt_n, sqrt_n)
            prompt = F.avg_pool2d(prompt, kernel_size=(stride, stride), stride=stride)
            prompt = prompt.reshape(num_img, c, -1).permute(0,2,1)
        return prompt


    def prepare_inputs_for_generation(self, input_ids, past_key_values=None,
                                      inputs_embeds=None, **kwargs):
        print("="*20, "prepare_inputs_for_generation", "="*20)
        images = kwargs.pop("images", None)
        image_sizes = kwargs.pop("image_sizes", None)
        inputs = super().prepare_inputs_for_generation(
            input_ids, past_key_values=past_key_values, inputs_embeds=inputs_embeds, **kwargs
        )
        if images is not None:
            inputs['images'] = images
        if image_sizes is not None:
            inputs['image_sizes'] = image_sizes
        return inputs

AutoConfig.register("blip3o_fast_inference", blip3oFastConfig)
AutoModelForCausalLM.register(blip3oFastConfig, blip3oFastForInferenceLM)