araison
/
explaining_framework
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fixing

This commit is contained in:
araison 2023-01-09 23:51:10 +01:00
parent e871f045a4
commit f7bb1e7858
3 changed files with 116 additions and 117 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

204
explaining_framework/utils/config_gen.py
View File

@ -4,14 +4,12 @@ import shutil
from explaining_framework.utils.io import read_yaml, write_yaml from explaining_framework.utils.io import read_yaml, write_yaml
from torch_geometric.data.makedirs import makedirs from torch_geometric.data.makedirs import makedirs
from torch_geometric.graphgym.loader import create_dataset
# from torch_geometric.graphgym.loader import load_pyg_dataset from torch_geometric.graphgym.utils.io import string_to_python
def chunkizing_list(l, n): def chunkizing_list(l, n):
for i in range(0, len(l), n): return [l[i : i + n] for i in range(0, len(l), n)]
yield l[i : i + n]
if "__main__" == __name__: if "__main__" == __name__:
@ -23,60 +21,61 @@ if "__main__" == __name__:
makedirs(explaining_folder) makedirs(explaining_folder)
explainer_folder = os.path.join(config_folder, "explaining") explainer_folder = os.path.join(config_folder, "explaining")
makedirs(explainer_folder) makedirs(explainer_folder)
chunk = list(range(154, 160, 2))
DATASET = [ DATASET = [
"CIFAR10", "CIFAR10",
"TRIANGLES", # "TRIANGLES",
"COLORS-3", # "COLORS-3",
"REDDIT-BINARY", # "REDDIT-BINARY",
"REDDIT-MULTI-5K", # "REDDIT-MULTI-5K",
"REDDIT-MULTI-12K", # "REDDIT-MULTI-12K",
"COLLAB", # "COLLAB",
"DBLP_v1", # "DBLP_v1",
"COIL-DEL", # "COIL-DEL",
"COIL-RAG", # "COIL-RAG",
"Fingerprint", # "Fingerprint",
"Letter-high", # "Letter-high",
"Letter-low", # "Letter-low",
"Letter-med", # "Letter-med",
"MSRC_9", # "MSRC_9",
"MSRC_21", # "MSRC_21",
"MSRC_21C", # "MSRC_21C",
"DD", # "DD",
"ENZYMES", # "ENZYMES",
"PROTEINS", # "PROTEINS",
"QM9", # "QM9",
"MUTAG", # "MUTAG",
"Mutagenicity", # "Mutagenicity",
"AIDS", # "AIDS",
"PATTERN", # "PATTERN",
"CLUSTER", # "CLUSTER",
"MNIST", # "MNIST",
"CIFAR10", # "CIFAR10",
"TSP", # "TSP",
"CSL", # "CSL",
"KarateClub", # "KarateClub",
"CS", # "CS",
"Physics", # "Physics",
"BBBP", # "BBBP",
"Tox21", # "Tox21",
"HIV", # "HIV",
"PCBA", # "PCBA",
"MUV", # "MUV",
"BACE", # "BACE",
"SIDER", # "SIDER",
"ClinTox", # "ClinTox",
"AIFB", # "AIFB",
"AM", # "AM",
"MUTAG", # "MUTAG",
"BGS", # "BGS",
"FAUST", # "FAUST",
"DynamicFAUST", # "DynamicFAUST",
"ShapeNet", # "ShapeNet",
"ModelNet10", # "ModelNet10",
"ModelNet40", # "ModelNet40",
"PascalVOC-SP", # "PascalVOC-SP",
"COCO-SP", # "COCO-SP",
] ]
EXPLAINER = [ EXPLAINER = [
"CAM", "CAM",
@ -95,57 +94,56 @@ if "__main__" == __name__:
"SCGNN", "SCGNN",
] ]
# for dataset_name in DATASET: for dataset_name in DATASET:
# try: for model_kind in ["best", "worst"]:
# dataset = load_pyg_dataset(name=dataset_name, dataset_dir="/tmp/") for explainer_name in EXPLAINER:
# except Exception as e: explaining_cfg = {}
# print(e) # explaining_cfg['adjust']['strategy']= 'rpns'
# continue # explaining_cfg['attack']['name']= 'all'
# for chunk in chunkizing_list(list(range(len(dataset))), 1000): explaining_cfg[
# for model_kind in ["best", "worst"]: "cfg_dest"
# for explainer_name in EXPLAINER: ] = f"dataset={dataset_name}-model={model_kind}=explainer={explainer_name}-chunk=[{chunk[0]},{chunk[-1]}]"
# explaining_cfg = {}
# # explaining_cfg['adjust']['strategy']= 'rpns' # = string_to_python(
# # explaining_cfg['attack']['name']= 'all' # f"dataset={dataset_name}-model={model_kind}=explainer={explainer_name}"
# explaining_cfg[ explaining_cfg["dataset"] = {}
# "cfg_dest" explaining_cfg["dataset"]["name"] = string_to_python(dataset_name)
# ] = f"dataset={dataset_name}-model={model_kind}=explainer={explainer_name}-chunk=[{chunk[0]},{chunk[-1]}]" explaining_cfg["dataset"]["item"] = chunk
# explaining_cfg["dataset"] = {} # explaining_cfg['explainer']['cfg']= 'default'
# explaining_cfg["dataset"]["name"] = dataset_name explaining_cfg["explainer"] = {}
# explaining_cfg["dataset"]["item"] = chunk explaining_cfg["explainer"]["name"] = string_to_python(explainer_name)
# # explaining_cfg['explainer']['cfg']= 'default' explaining_cfg["explainer"]["force"] = True
# explaining_cfg["explainer"] = {} explaining_cfg["explanation_type"] = string_to_python("phenomenon")
# explaining_cfg["explainer"]["name"] = explainer_name # explaining_cfg['metrics']['accuracy']['name']='all'
# explaining_cfg["explanation_type"] = "phenomenon" # explaining_cfg['metrics']['fidelity']['name']='all'
# # explaining_cfg['metrics']['accuracy']['name']='all' # explaining_cfg['metrics']['sparsity']['name']='all'
# # explaining_cfg['metrics']['fidelity']['name']='all' explaining_cfg["model"] = {}
# # explaining_cfg['metrics']['sparsity']['name']='all' explaining_cfg["model"]["ckpt"] = string_to_python(model_kind)
# explaining_cfg["model"] = {} explaining_cfg["model"]["path"] = string_to_python(
# explaining_cfg["model"]["ckpt"] = model_kind "/home/SIC/araison/test_ggym/pytorch_geometric/graphgym/results"
# explaining_cfg["model"][ )
# "path" # explaining_cfg['out_dir']='./explanation'
# ] = "/home/SIC/araison/test_ggym/pytorch_geometric/graphgym/results" # explaining_cfg['print']='both'
# # explaining_cfg['out_dir']='./explanation' # explaining_cfg['threshold']['config']['type']='all'
# # explaining_cfg['print']='both' # explaining_cfg['threshold']['value']['hard']=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
# # explaining_cfg['threshold']['config']['type']='all' # explaining_cfg['threshold']['value']['topk']=[2, 3, 5, 10, 20, 30, 50]
# # explaining_cfg['threshold']['value']['hard']=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9] PATH = os.path.join(
# # explaining_cfg['threshold']['value']['topk']=[2, 3, 5, 10, 20, 30, 50] explaining_folder + "/" + explaining_cfg["cfg_dest"] + ".yaml",
# PATH = os.path.join( )
# explaining_folder + "/" + explaining_cfg["cfg_dest"] + ".yaml", write_yaml(explaining_cfg, PATH)
# )
# if os.path.exists(PATH): # if os.path.exists(PATH):
# continue # continue
# else: # else:
# write_yaml(explaining_cfg, PATH) # write_yaml(explaining_cfg, PATH)
configs = [ # configs = [
path for path in glob.glob(os.path.join(explaining_folder, "**", "*.yaml")) # path for path in glob.glob(os.path.join(explaining_folder, "**", "*.yaml"))
] # ]
for path in configs: # for path in configs:
data = read_yaml(path) # data = read_yaml(path)
data["model"][ # data["model"][
"path" # "path"
] = "/media/data/SIC/araison/exps/pyg_fork/graphgym/results/graph_classif_base_grid_graph_classif_grid" # ] = "/media/data/SIC/araison/exps/pyg_fork/graphgym/results/graph_classif_base_grid_graph_classif_grid"
write_yaml(data, path) # write_yaml(data, path)
# for index, config_chunk in enumerate( # for index, config_chunk in enumerate(
# chunkizing_list(configs, int(len(configs) / 5)) # chunkizing_list(configs, int(len(configs) / 5))

27
explaining_framework/utils/explaining/outline.py
View File

@ -6,6 +6,20 @@ import os
from typing import Any from typing import Any
from eixgnn.eixgnn import EiXGNN from eixgnn.eixgnn import EiXGNN
from scgnn.scgnn import SCGNN
from torch_geometric import seed_everything
from torch_geometric.data import Batch, Data
from torch_geometric.data.makedirs import makedirs
from torch_geometric.explain import Explainer
from torch_geometric.explain.config import ThresholdConfig
from torch_geometric.explain.explanation import Explanation
from torch_geometric.graphgym.config import cfg
from torch_geometric.graphgym.loader import create_dataset
from torch_geometric.graphgym.model_builder import cfg, create_model
from torch_geometric.graphgym.utils.device import auto_select_device
from torch_geometric.loader.dataloader import DataLoader
from yacs.config import CfgNode as CN
from explaining_framework.config.explainer_config.eixgnn_config import \ from explaining_framework.config.explainer_config.eixgnn_config import \
eixgnn_cfg eixgnn_cfg
from explaining_framework.config.explainer_config.scgnn_config import scgnn_cfg from explaining_framework.config.explainer_config.scgnn_config import scgnn_cfg
@ -31,19 +45,6 @@ from explaining_framework.utils.io import (dump_cfg, is_exists,
obj_config_to_str, read_json, obj_config_to_str, read_json,
set_printing, write_json, set_printing, write_json,
write_yaml) write_yaml)
from scgnn.scgnn import SCGNN
from torch_geometric import seed_everything
from torch_geometric.data import Batch, Data
from torch_geometric.data.makedirs import makedirs
from torch_geometric.explain import Explainer
from torch_geometric.explain.config import ThresholdConfig
from torch_geometric.explain.explanation import Explanation
from torch_geometric.graphgym.config import cfg
from torch_geometric.graphgym.loader import create_dataset
from torch_geometric.graphgym.model_builder import cfg, create_model
from torch_geometric.graphgym.utils.device import auto_select_device
from torch_geometric.loader.dataloader import DataLoader
from yacs.config import CfgNode as CN
all__captum = [ all__captum = [
"LRP", "LRP",

2
explaining_framework/utils/io.py
View File

@ -26,7 +26,7 @@ def read_yaml(path: str) -> dict:
def write_yaml(data: dict, path: str) -> None: def write_yaml(data: dict, path: str) -> None:
with open(path, "w") as f: with open(path, "w") as f:
data = yaml.dump(data, f) data = yaml.safe_dump(data, f, default_flow_style=False)
def dump_cfg(cfg, path): def dump_cfg(cfg, path):