new_impl/cv/sam/seg_wo_fbs.py

import torch
import sys
from torch import nn
from new_impl.cv.elasticdnn.api.model import ElasticDNN_OfflineSegFMModel, ElasticDNN_OfflineSegMDModel
from new_impl.cv.elasticdnn.api.algs.md_pretraining_wo_fbs import ElasticDNN_MDPretrainingWoFBSAlg
from new_impl.cv.elasticdnn.model.base import ElasticDNNUtil
from new_impl.cv.elasticdnn.pipeline.offline.fm_to_md.base import FM_to_MD_Util
from sam import FM_to_MD_sam_Util
from new_impl.cv.elasticdnn.pipeline.offline.fm_lora.base import FMLoRA_Util
from sam import FMLoRA_sam_Util
from sam import ElasticsamUtil
from utils.dl.common.model import LayerActivation, get_module, get_parameter
from utils.common.exp import save_models_dict_for_init, get_res_save_dir
from data import build_scenario
from utils.dl.common.loss import CrossEntropyLossSoft
import torch.nn.functional as F


class ElasticDNN_ViT_OfflineSegFMModel(ElasticDNN_OfflineSegFMModel):
    def generate_md_by_reducing_width(self, reducing_width_ratio, samples: torch.Tensor):
        return FM_to_MD_sam_Util().init_md_from_fm_by_reducing_width_with_perf_test(self.models_dict['main'], 
                                                                        reducing_width_ratio, samples).to(self.device)
        
    def get_feature_hook(self) -> LayerActivation:
        return LayerActivation(get_module(self.models_dict['main'], 'head'), True, self.device)
    
    def get_elastic_dnn_util(self) -> ElasticDNNUtil:
        return ElasticsamUtil()
    
    def forward_to_get_task_loss(self, x, y, *args, **kwargs):
        return F.cross_entropy(self.infer(x), y)
    
    def get_lora_util(self) -> FMLoRA_Util:
        return FMLoRA_sam_Util()
    
    def get_task_head_params(self):
        head = get_module(self.models_dict['main'], 'head')
        return list(head.parameters())
        
        
class ElasticDNN_ViT_OfflineSegMDModel(ElasticDNN_OfflineSegMDModel):
        
    def get_feature_hook(self) -> LayerActivation:
        return LayerActivation(get_module(self.models_dict['main'], 'head'), True, self.device)
    
    def forward_to_get_task_loss(self, x, y, *args, **kwargs):
        return F.cross_entropy(self.infer(x), y)
    
    def get_distill_loss(self, student_output, teacher_output):
        return F.mse_loss(student_output, teacher_output)
    
    def get_matched_param_of_fm(self, self_param_name, fm: nn.Module):
        if any([k in self_param_name for k in ['fbs', 'cls_token', 'pos_embed']]):
            return None
        
        # 1. xx.qkv.to_qkv.yy to xx.qkv.qkv.aa and xx.qkv.abs.zz
        if 'to_qkv.weight' in self_param_name:
            ss = self_param_name.split('.')
            
            fm_qkv_name = '.'.join(ss[0: -2]) + '.fc'
            fm_qkv = get_module(fm, fm_qkv_name)
            
            fm_abs_name = '.'.join(ss[0: -2]) + '.ab'
            fm_abs = get_module(fm, fm_abs_name)
            
            return torch.cat([
                fm_qkv.weight.data, # task-agnositc params
                torch.cat([(_abs[0].weight.T @ _abs[1].weight.T).T for _abs in fm_abs], dim=0) # task-specific params (LoRA)
            ], dim=0)
            
        elif 'to_qkv.bias' in self_param_name:
            ss = self_param_name.split('.')
            
            fm_qkv_name = '.'.join(ss[0: -2]) + '.qkv.bias'
            return get_parameter(fm, fm_qkv_name)
            
        elif 'mlp.fc1' in self_param_name:
            fm_param_name = self_param_name.replace('.linear', '')
            return get_parameter(fm, fm_param_name)

        else:
            return get_parameter(fm, self_param_name)
        
        
if __name__ == '__main__':
    from utils.dl.common.env import set_random_seed
    set_random_seed(1)
    
    # 3. init scenario
    scenario = build_scenario(
        source_datasets_name=['GTA5', 'SuperviselyPerson'],
        target_datasets_order=['Cityscapes', 'BaiduPerson'] * 10,
        da_mode='close_set',
        data_dirs={
            'GTA5': '/data/zql/datasets/GTA-ls-copy/GTA5',
            'SuperviselyPerson': '/data/zql/datasets/supervisely_person/Supervisely Person Dataset',
            'Cityscapes': '/data/zql/datasets/cityscape/',
            'BaiduPerson': '/data/zql/datasets/baidu_person/clean_images/'
        },
    )
    
    # 1. init model
    # from dnns.deeplabv3.head import modify_forward_head
    # modify_forward_head() # TODO: bring a bug
    from dnns.vit import vit_b_16
    fm_models_dict_path = 'new_impl/cv/sam/results/seg.py/20231123/999983-212616/models/fm_best.pt'
    fm_models = torch.load(fm_models_dict_path)
    # for n,m in fm_models['main'].named_modules():
    #     print(n)
    # from utils.dl.common.model import set_module
    # set_module(
    #     fm_models['main'],
    #     'norm',
    #     nn.Sequential(
    #         get_module(fm_models['main'], 'norm'),
    #         get_module(fm_models['main'], 'head')
    #     )
    # )
    # set_module(fm_models['main'], 'head', nn.Identity())
    # fm_models['main'].forward = fm_models['main'].forward_features
    
    fm_models_dict_path = save_models_dict_for_init(fm_models, __file__, 'fm_sam_seg_lora')
    md_models_dict_path = save_models_dict_for_init({
        'main': -1
    }, __file__, 'md_sam_none')
    device = 'cuda'
    
    fm_model = ElasticDNN_ViT_OfflineSegFMModel('fm', fm_models_dict_path, device, scenario.num_classes)
    md_model = ElasticDNN_ViT_OfflineSegMDModel('md', md_models_dict_path, device, scenario.num_classes)
    
    # 2. init alg
    models = {
        'fm': fm_model,
        'md': md_model
    }
    fm_to_md_alg = ElasticDNN_MDPretrainingWoFBSAlg(models, get_res_save_dir(__file__, None))
    
    from utils.dl.common.lr_scheduler import get_linear_schedule_with_warmup
    fm_to_md_alg.run(scenario, hyps={
        'launch_tbboard': False,
        
        'samples_size': (1, 3, 224, 224),
        'generate_md_width_ratio': 8,
        
        'train_batch_size': 16,
        'val_batch_size': 128,
        'num_workers': 16,
        'optimizer': 'AdamW',
        'optimizer_args': {'lr': 5e-4, 'betas': [0.9, 0.999], 'weight_decay': 0.01},
        'scheduler': 'LambdaLR',
        'scheduler_args': {'lr_lambda': get_linear_schedule_with_warmup(10000, 70000)},
        'num_iters': 80000,
        'val_freq': 1000,
        'distill_loss_weight': 1.0
    })