Code to ensemble multiple models

ensemble.py

try:
    # ignore ShapelyDeprecationWarning from fvcore
    import warnings

    from shapely.errors import ShapelyDeprecationWarning

    warnings.filterwarnings("ignore", category=ShapelyDeprecationWarning)
except:
    pass
import copy
import itertools
import logging
import os
from collections import OrderedDict, defaultdict
from typing import Any, Dict, List, Set

import detectron2.utils.comm as comm
import torch
from detectron2.checkpoint import DetectionCheckpointer
from detectron2.config import get_cfg
from detectron2.data import MetadataCatalog
from detectron2.engine import (
    DefaultTrainer,
    default_argument_parser,
    default_setup,
    launch,
)
from detectron2.evaluation import (
    CityscapesInstanceEvaluator,
    CityscapesSemSegEvaluator,
    COCOEvaluator,
    COCOPanopticEvaluator,
    DatasetEvaluators,
    LVISEvaluator,
    SemSegEvaluator,
    verify_results,
)
from torch.nn.parallel import DistributedDataParallel
from detectron2.projects.deeplab import add_deeplab_config, build_lr_scheduler
from detectron2.solver.build import maybe_add_gradient_clipping
from detectron2.utils.logger import setup_logger
from tabulate import tabulate
from termcolor import colored
from san.models import layers
from san import (
    COCOInstanceNewBaselineDatasetMapper,
    COCOPanopticNewBaselineDatasetMapper,
    InstanceSegEvaluator,
    MaskFormerInstanceDatasetMapper,
    MaskFormerPanopticDatasetMapper,
    MaskFormerSemanticDatasetMapper,
    SemanticSegmentorWithTTA,
    add_san_config,
)
from san.data import build_detection_test_loader, build_detection_train_loader
from san.utils import WandbWriter, setup_wandb
from san.utils.hooks import ModelAfterStepHook
from torch import nn


class ModelEnsemble(nn.Module):
    """
    A SemanticSegmentor with test-time augmentation enabled.
    Its :meth:`__call__` method has the same interface as :meth:`SemanticSegmentor.forward`.
    """

    def __init__(self, models):
        """
        Args:
            cfg (CfgNode):
            model (SemanticSegmentor): a SemanticSegmentor to apply TTA on.
            tta_mapper (callable): takes a dataset dict and returns a list of
                augmented versions of the dataset dict. Defaults to
                `DatasetMapperTTA(cfg)`.
            batch_size (int): batch the augmented images into this batch size for inference.
        """
        super().__init__()
        self.models = nn.ModuleList()
        for model in models:
            if isinstance(model, DistributedDataParallel):
                model = model.module
            self.models.append(model)

    def __call__(self, batched_inputs):
        """
        Same input/output format as :meth:`SemanticSegmentor.forward`
        """
        results = []
        for model in self.models:
            results.append(model(batched_inputs))

        final_results = []
        for predictions in zip(*results):
            sem_segs = [pred["sem_seg"].cpu() for pred in predictions]
            min_cls = min([sem_seg.shape[0] for sem_seg in sem_segs])
            assert len(sem_segs)==2
            w = float(os.environ["EWEIGHT"])
            sem_seg = torch.pow(sem_segs[0][:min_cls],w)*torch.pow(sem_segs[1][:min_cls],1-w)
            # sem_seg = torch.stack(sem_segs, dim=0).mean(dim=0)
            torch.cuda.empty_cache()
            final_results.append({"sem_seg": sem_seg})

        return final_results


class Trainer(DefaultTrainer):
    """
    Extension of the Trainer class adapted to MaskFormer.
    """

    def build_hooks(self):
        rets = super().build_hooks()
        rets.append(ModelAfterStepHook(self.model))
        return rets

    def build_writers(self):
        """
        Build a list of writers to be used. By default it contains
        writers that write metrics to the screen,
        a json file, and a tensorboard event file respectively.
        If you'd like a different list of writers, you can overwrite it in
        your trainer.

        Returns:
            list[EventWriter]: a list of :class:`EventWriter` objects.

        It is now implemented by:
        ::
            return [
                CommonMetricPrinter(self.max_iter),
                JSONWriter(os.path.join(self.cfg.OUTPUT_DIR, "metrics.json")),
                TensorboardXWriter(self.cfg.OUTPUT_DIR),
            ]

        """
        writers = super().build_writers()
        writers[-1] = WandbWriter()
        return writers

    @classmethod
    def build_evaluator(cls, cfg, dataset_name, output_folder=None):
        """
        Create evaluator(s) for a given dataset.
        This uses the special metadata "evaluator_type" associated with each
        builtin dataset. For your own dataset, you can simply create an
        evaluator manually in your script and do not have to worry about the
        hacky if-else logic here.
        """
        if output_folder is None:
            output_folder = os.path.join(cfg.OUTPUT_DIR, "inference")
        evaluator_list = []
        evaluator_type = MetadataCatalog.get(dataset_name).evaluator_type
        # semantic segmentation
        if evaluator_type in ["sem_seg", "ade20k_panoptic_seg"]:
            evaluator_list.append(
                SemSegEvaluator(
                    dataset_name,
                    distributed=True,
                    output_dir=output_folder,
                )
            )
        # instance segmentation
        if evaluator_type == "coco":
            evaluator_list.append(COCOEvaluator(dataset_name, output_dir=output_folder))
        # panoptic segmentation
        if evaluator_type in [
            "coco_panoptic_seg",
            "ade20k_panoptic_seg",
            "cityscapes_panoptic_seg",
            "mapillary_vistas_panoptic_seg",
        ]:
            if cfg.MODEL.MASK_FORMER.TEST.PANOPTIC_ON:
                evaluator_list.append(
                    COCOPanopticEvaluator(dataset_name, output_folder)
                )
        # COCO
        if (
            evaluator_type == "coco_panoptic_seg"
            and cfg.MODEL.MASK_FORMER.TEST.INSTANCE_ON
        ):
            evaluator_list.append(COCOEvaluator(dataset_name, output_dir=output_folder))
        if (
            evaluator_type == "coco_panoptic_seg"
            and cfg.MODEL.MASK_FORMER.TEST.SEMANTIC_ON
        ):
            evaluator_list.append(
                SemSegEvaluator(
                    dataset_name, distributed=True, output_dir=output_folder
                )
            )
        # Mapillary Vistas
        if (
            evaluator_type == "mapillary_vistas_panoptic_seg"
            and cfg.MODEL.MASK_FORMER.TEST.INSTANCE_ON
        ):
            evaluator_list.append(
                InstanceSegEvaluator(dataset_name, output_dir=output_folder)
            )
        if (
            evaluator_type == "mapillary_vistas_panoptic_seg"
            and cfg.MODEL.MASK_FORMER.TEST.SEMANTIC_ON
        ):
            evaluator_list.append(
                SemSegEvaluator(
                    dataset_name, distributed=True, output_dir=output_folder
                )
            )
        # Cityscapes
        if evaluator_type == "cityscapes_instance":
            assert (
                torch.cuda.device_count() > comm.get_rank()
            ), "CityscapesEvaluator currently do not work with multiple machines."
            return CityscapesInstanceEvaluator(dataset_name)
        if evaluator_type == "cityscapes_sem_seg":
            assert (
                torch.cuda.device_count() > comm.get_rank()
            ), "CityscapesEvaluator currently do not work with multiple machines."
            return CityscapesSemSegEvaluator(dataset_name)
        if evaluator_type == "cityscapes_panoptic_seg":
            if cfg.MODEL.MASK_FORMER.TEST.SEMANTIC_ON:
                assert (
                    torch.cuda.device_count() > comm.get_rank()
                ), "CityscapesEvaluator currently do not work with multiple machines."
                evaluator_list.append(CityscapesSemSegEvaluator(dataset_name))
            if cfg.MODEL.MASK_FORMER.TEST.INSTANCE_ON:
                assert (
                    torch.cuda.device_count() > comm.get_rank()
                ), "CityscapesEvaluator currently do not work with multiple machines."
                evaluator_list.append(CityscapesInstanceEvaluator(dataset_name))
        # ADE20K
        if (
            evaluator_type == "ade20k_panoptic_seg"
            and cfg.MODEL.MASK_FORMER.TEST.INSTANCE_ON
        ):
            evaluator_list.append(
                InstanceSegEvaluator(dataset_name, output_dir=output_folder)
            )
        # LVIS
        if evaluator_type == "lvis":
            return LVISEvaluator(dataset_name, output_dir=output_folder)
        if len(evaluator_list) == 0:
            raise NotImplementedError(
                "no Evaluator for the dataset {} with the type {}".format(
                    dataset_name, evaluator_type
                )
            )
        elif len(evaluator_list) == 1:
            return evaluator_list[0]
        return DatasetEvaluators(evaluator_list)

    @classmethod
    def build_train_loader(cls, cfg):
        # Semantic segmentation dataset mapper
        if cfg.INPUT.DATASET_MAPPER_NAME == "mask_former_semantic":
            mapper = MaskFormerSemanticDatasetMapper(cfg, True)
            return build_detection_train_loader(cfg, mapper=mapper)
        # Panoptic segmentation dataset mapper
        elif cfg.INPUT.DATASET_MAPPER_NAME == "mask_former_panoptic":
            mapper = MaskFormerPanopticDatasetMapper(cfg, True)
            return build_detection_train_loader(cfg, mapper=mapper)
        # Instance segmentation dataset mapper
        elif cfg.INPUT.DATASET_MAPPER_NAME == "mask_former_instance":
            mapper = MaskFormerInstanceDatasetMapper(cfg, True)
            return build_detection_train_loader(cfg, mapper=mapper)
        # coco instance segmentation lsj new baseline
        elif cfg.INPUT.DATASET_MAPPER_NAME == "coco_instance_lsj":
            mapper = COCOInstanceNewBaselineDatasetMapper(cfg, True)
            return build_detection_train_loader(cfg, mapper=mapper)
        # coco panoptic segmentation lsj new baseline
        elif cfg.INPUT.DATASET_MAPPER_NAME == "coco_panoptic_lsj":
            mapper = COCOPanopticNewBaselineDatasetMapper(cfg, True)
            return build_detection_train_loader(cfg, mapper=mapper)
        else:
            mapper = None
            return build_detection_train_loader(cfg, mapper=mapper)

    @classmethod
    def build_test_loader(cls, cfg, dataset_name):
        """
        Returns:
            iterable

        It now calls :func:`detectron2.data.build_detection_test_loader`.
        Overwrite it if you'd like a different data loader.
        """
        return build_detection_test_loader(cfg, dataset_name)

    @classmethod
    def build_lr_scheduler(cls, cfg, optimizer):
        """
        It now calls :func:`detectron2.solver.build_lr_scheduler`.
        Overwrite it if you'd like a different scheduler.
        """
        return build_lr_scheduler(cfg, optimizer)

    @classmethod
    def build_optimizer(cls, cfg, model):
        weight_decay_norm = cfg.SOLVER.WEIGHT_DECAY_NORM
        weight_decay_embed = cfg.SOLVER.WEIGHT_DECAY_EMBED

        defaults = {}
        defaults["lr"] = cfg.SOLVER.BASE_LR
        defaults["weight_decay"] = cfg.SOLVER.WEIGHT_DECAY

        norm_module_types = (
            torch.nn.BatchNorm1d,
            torch.nn.BatchNorm2d,
            torch.nn.BatchNorm3d,
            torch.nn.SyncBatchNorm,
            # NaiveSyncBatchNorm inherits from BatchNorm2d
            torch.nn.GroupNorm,
            torch.nn.InstanceNorm1d,
            torch.nn.InstanceNorm2d,
            torch.nn.InstanceNorm3d,
            torch.nn.LayerNorm,
            torch.nn.LocalResponseNorm,
            layers.LayerNorm,
        )

        params: List[Dict[str, Any]] = []
        memo: Set[torch.nn.parameter.Parameter] = set()
        for module_name, module in model.named_modules():
            for module_param_name, value in module.named_parameters(recurse=False):
                if not value.requires_grad:
                    continue
                # Avoid duplicating parameters
                if value in memo:
                    continue
                memo.add(value)

                hyperparams = copy.copy(defaults)
                hyperparams["param_name"] = ".".join([module_name, module_param_name])
                if "backbone" in module_name:
                    hyperparams["lr"] = (
                        hyperparams["lr"] * cfg.SOLVER.BACKBONE_MULTIPLIER
                    )
                if "masked_pooling_layer" in module_name:
                    hyperparams["lr"] = (
                        hyperparams["lr"] * cfg.SOLVER.CLIPHEAD_MULTIPLIER
                    )
                if (
                    "relative_position_bias_table" in module_param_name
                    or "absolute_pos_embed" in module_param_name
                    or "positional_embedding" in module_param_name
                    or "pos_embed" in module_param_name
                    or "query_token" in module_param_name
                ):
                    print(module_param_name)
                    hyperparams["weight_decay"] = 0.0
                if isinstance(module, norm_module_types):
                    hyperparams["weight_decay"] = weight_decay_norm
                if isinstance(module, torch.nn.Embedding):
                    hyperparams["weight_decay"] = weight_decay_embed
                params.append({"params": [value], **hyperparams})

        def maybe_add_full_model_gradient_clipping(optim):
            # detectron2 doesn't have full model gradient clipping now
            clip_norm_val = cfg.SOLVER.CLIP_GRADIENTS.CLIP_VALUE
            enable = (
                cfg.SOLVER.CLIP_GRADIENTS.ENABLED
                and cfg.SOLVER.CLIP_GRADIENTS.CLIP_TYPE == "full_model"
                and clip_norm_val > 0.0
            )

            class FullModelGradientClippingOptimizer(optim):
                def step(self, closure=None):
                    all_params = itertools.chain(
                        *[x["params"] for x in self.param_groups]
                    )
                    torch.nn.utils.clip_grad_norm_(all_params, clip_norm_val)
                    super().step(closure=closure)

            return FullModelGradientClippingOptimizer if enable else optim

        optimizer_type = cfg.SOLVER.OPTIMIZER
        if optimizer_type == "SGD":
            optimizer = maybe_add_full_model_gradient_clipping(torch.optim.SGD)(
                params, cfg.SOLVER.BASE_LR, momentum=cfg.SOLVER.MOMENTUM
            )
        elif optimizer_type == "ADAMW":
            optimizer = maybe_add_full_model_gradient_clipping(torch.optim.AdamW)(
                params, cfg.SOLVER.BASE_LR
            )
        else:
            raise NotImplementedError(f"no optimizer type {optimizer_type}")
        if not cfg.SOLVER.CLIP_GRADIENTS.CLIP_TYPE == "full_model":
            optimizer = maybe_add_gradient_clipping(cfg, optimizer)
        # display the lr and wd of each param group in a table
        optim_info = defaultdict(list)
        total_params_size = 0
        for group in optimizer.param_groups:
            optim_info["Param Name"].append(group["param_name"])
            optim_info["Param Shape"].append(
                "X".join([str(x) for x in list(group["params"][0].shape)])
            )
            total_params_size += group["params"][0].numel()
            optim_info["Lr"].append(group["lr"])
            optim_info["Wd"].append(group["weight_decay"])
        # Counting the number of parameters
        optim_info["Param Name"].append("Total")
        optim_info["Param Shape"].append(
            "{:.2f}M".format(total_params_size / 1024 / 1024)
        )
        optim_info["Lr"].append("-")
        optim_info["Wd"].append("-")
        table = tabulate(
            list(zip(*optim_info.values())),
            headers=optim_info.keys(),
            tablefmt="grid",
            floatfmt=".2e",
            stralign="center",
            numalign="center",
        )
        logger = logging.getLogger("e2e")
        logger.info("Optimizer Info:\n{}\n".format(table))
        return optimizer

    @classmethod
    def test_with_TTA(cls, cfg, model):
        logger = logging.getLogger("detectron2.trainer")
        # In the end of training, run an evaluation with TTA.
        logger.info("Running inference with test-time augmentation ...")
        model = SemanticSegmentorWithTTA(cfg, model)
        evaluators = [
            cls.build_evaluator(
                cfg, name, output_folder=os.path.join(cfg.OUTPUT_DIR, "inference_TTA")
            )
            for name in cfg.DATASETS.TEST
        ]
        res = cls.test(cfg, model, evaluators)
        res = OrderedDict({k + "_TTA": v for k, v in res.items()})
        return res


def setup(args, logger=False):
    """
    Create configs and perform basic setups.
    """
    cfg = get_cfg()
    # for poly lr schedule
    add_deeplab_config(cfg)
    add_san_config(cfg)
    cfg.merge_from_file(args.config_file)
    cfg.merge_from_list(args.opts)
    cfg.freeze()
    default_setup(cfg, args)
    # Setup logger for "mask_former" module
    if logger:
        setup_logger(
            output=cfg.OUTPUT_DIR, distributed_rank=comm.get_rank(), name="e2e"
        )
    return cfg


def main(args_list):
    models = []
    for args in args_list:
        cfg = setup(args, logger=len(models) == 0)
        model = Trainer.build_model(cfg)
        DetectionCheckpointer(model, save_dir=cfg.OUTPUT_DIR).resume_or_load(
            cfg.MODEL.WEIGHTS, resume=args.resume
        )
        models.append(model)

    model = ModelEnsemble(models)
    res = Trainer.test(cfg, model)


if __name__ == "__main__":
    import sys
    args_file = sys.argv[1]
    if len(sys.argv) >2:
        extra_opt = sys.argv[2:]
    else:
        extra_opt = []
    with open(args_file, "r") as f:
        args_list = f.readlines()
        args_list = [
            ["--eval-only"] + x.split(" ")+extra_opt
            for x in args_list
            if not x.strip().startswith("#")
        ]
    # extend args
    if "{run}" in ' '.join([' '.join(args) for args in args_list]):
        args_list_group = []
        for run_id in range(5):
            args_list_group.append(
                [
                    [x.replace("{run}", str(run_id)) for x in args]
                    for args in args_list
                ]
            )
    else:
        args_list_group = [args_list]
    print("Will run {} experiments".format(len(args_list_group)))
    for args_list in args_list_group:
        print("=====================================")
        print(args_list)
    print("=====================================")
    for _args_list in args_list_group:
        args_list = [default_argument_parser().parse_args(args) for args in _args_list]
        print("Command Line Args:", args_list)
        args = args_list[0]
        launch(
            main,
            8,
            num_machines=args.num_machines,
            machine_rank=args.machine_rank,
            dist_url=args.dist_url,
            args=(args_list,),
        )