explaining_framework/main.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
#

import os
import time

from torch_geometric import seed_everything
from torch_geometric.data.makedirs import makedirs
from torch_geometric.explain import Explainer
from torch_geometric.explain.config import ThresholdConfig
from torch_geometric.graphgym.config import cfg
from torch_geometric.graphgym.utils.device import auto_select_device

from explaining_framework.config.explaining_config import explaining_cfg
from explaining_framework.utils.explaining.cmd_args import parse_args
from explaining_framework.utils.explaining.outline import ExplainingOutline
from explaining_framework.utils.explanation.adjust import Adjust
from explaining_framework.utils.io import (obj_config_to_str, read_json,
                                           write_json, write_yaml)

# inference, time, force,


def get_pred(explanation, force=False):
    dict_ = explanation.to_dict()
    if dict_.get("pred") is None or dict_.get("pred_masked") or force:
        pred = explainer.get_prediction(explanation)
        pred_masked = explainer.get_masked_prediction(
            x=explanation.x,
            edge_index=explanation.edge_index,
            node_mask=explanation.node_mask,
            edge_mask=explanation.edge_mask,
        )
        explanation.__setattr__("pred", pred)
        explanation.__setattr__("pred_masked", pred_masked)
        return explanation
    else:
        return explanation


if __name__ == "__main__":
    args = parse_args()
    outline = ExplainingOutline(args.explaining_cfg_file)
    auto_select_device()

    # Load components
    dataset = outline.dataset.to(cfg.accelerator)
    model = outline.model.to(cfg.accelerator)
    model_info = outline.model_info
    metrics = outline.metrics
    explaining_algorithm = outline.explaining_algorithm
    attacks = outline.attacks
    explainer_cfg = outline.explainer_cfg
    model_signature = outline.model_signature

    # Set seed
    seed_everything(explaining_cfg.seed)

    # Global path
    global_path = os.path.join(explaining_cfg.out_dir, model_signature)
    makedirs(global_path)
    write_yaml(cfg, os.path.join(global_path, "config.yaml"))
    write_json(model_info, os.path.join(global_path, "info.json"))

    global_path = os.path.join(
        global_path,
        explaining_cfg.explainer.name + "_" + obj_config_to_str(explaining_algorithm),
    )
    makedirs(global_path)
    write_yaml(explaining_cfg, os.path.join(global_path, explaining_cfg.cfg_dest))
    write_yaml(explainer_cfg, os.path.join(global_path, "explainer_cfg.yaml"))

    global_path = os.path.join(global_path, obj_config_to_str(explaining_algorithm))
    makedirs(global_path)
    explainer = Explainer(
        model=model,
        algorithm=explaining_algorithm,
        explainer_config=dict(
            explanation_type=explaining_cfg.explanation_type,
            node_mask_type="object",
            edge_mask_type="object",
        ),
        model_config=dict(
            mode="regression",
            task_level=cfg.dataset.task,
            return_type=explaining_cfg.model_config.return_type,
        ),
    )
    # Save explaining configuration
    for index, item in enumerate(dataset):
        save_raw_path = os.path.join(global_path, "raw")
        makedirs(save_raw_path)
        explanation_path = os.path.join(save_raw_path, f"{index}.json")

        if is_exists(explanation_path):
            if explaining_cfg.explainer.force:
                explanation = explainer(
                    x=item.x,
                    edge_index=item.edge_index,
                    index=item.y,
                    target=item.y,
                )
            else:
                explanation = load_explanation(explanation_path)
        else:
            explanation = explainer(
                x=item.x,
                edge_index=item.edge_index,
                index=item.y,
                target=item.y,
            )
        explanation = get_pred(explanation, force=False)
        save_explanation(explanation, explanation_path)
        for apply_relu in [True, False]:
            for apply_absolute in [True, False]:
                adjust = Adjust(apply_relu=apply_relu, apply_absolute=apply_absolute)
                save_raw_path = os.path.join(
                    global_path, f"adjust-{obj_config_to_str(adjust)}"
                )
                makedirs(save_raw_path)
                explanation = adjust.forward(explanation)
                explanation_path = os.path.join(save_raw_path, f"{index}.json")
                explanation = get_pred(explanation, force=True)
                save_explanation(explanation, explanation_path)

                for threshold_approach in ["hard", "topk", "topk_hard"]:
                    for threshold_value in explaining_cfg.threshold_config.value:

                        masking_path = os.path.join(
                            save_raw_path,
                            f"threshold={threshold_approach}-value={value}",
                        )
                        exp_threshold_path = os.path.join(masking_path, f"{index}.json")
                        if is_exists(exp_threshold_path):
                            explanation = load_explanation(exp_threshold_path)
                        else:
                            threshold_conf = {
                                "threshold_type": threshold_approach,
                                "value": threshold_value,
                            }
                            explainer.threshold_config = ThresholdConfig.cast(
                                threshold_conf
                            )

                            expl = copy.copy(explanation)
                            exp_threshold = explainer._post_process(expl)
                            exp_threshold = get_pred(exp_threshold, force=True)

                            save_explanation(exp_threshold, exp_threshold_path)
                        for metric in metrics:
                            metric_path = os.path.join(
                                masking_path, f"{obj_config_to_str(metric)}"
                            )
                            if is_exists(os.path.join(metric_path, f"{index}.json")):
                                continue
                            else:
                                out = metric.forward(exp_threshold)
                                write_json({f"{metric.name}": out})