isort.

Author: janfb

notebooks/mnle_utils.py

@@ -2,11 +2,9 @@
 
 import numpy as np
 import torch
-
-from torch import Tensor, nn
 from sbi.utils.sbiutils import standardizing_net
+from torch import Tensor, nn
 from torch.distributions import Bernoulli
-from torch import Tensor
 
 
 def build_choice_net(

notebooks/revision-experiments/lan_generate_data.py

@@ -4,10 +4,10 @@
 
 # Load necessary packages
 from copy import deepcopy
-import torch
 
 import sbibm
 import ssms
+import torch
 
 # Get benchmark task to load Julia simulator.
 seed = torch.randint(100000, (1,)).item()

notebooks/revision-experiments/lan_run_inference.py

@@ -3,14 +3,13 @@
 # function wrapper in utils.
 
 import pickle
-from pathlib import Path
-from joblib import Parallel, delayed
 import sys
+from pathlib import Path
 
 import lanfactory
 import sbibm
 import torch
-
+from joblib import Parallel, delayed
 from sbi.inference import MCMCPosterior
 from sbi.utils import mcmc_transform
 

notebooks/revision-experiments/lan_run_training.py

@@ -1,15 +1,14 @@
 # Adapted from: https://github.com/AlexanderFengler/LANfactory
 # Script to run LAN training given a pre-simulated data set.
 
-# Load necessary packages
-import lanfactory
 import os
 from copy import deepcopy
-import torch
-
-from lanfactory.trainers import ModelTrainerTorchMLP
 from pathlib import Path
 
+# Load necessary packages
+import lanfactory
+import torch
+from lanfactory.trainers import ModelTrainerTorchMLP
 
 BASE_DIR = Path.cwd()
 

notebooks/revision-experiments/mnle_run_inference.py

@@ -5,7 +5,6 @@
 
 import sbibm
 import torch
-
 from sbi.inference import MNLE
 
 # Get benchmark task to load observations

notebooks/revision-experiments/mnle_training_script.py

@@ -1,14 +1,13 @@
 # Script for training MNLE with pre-simulated data.
 
 import pickle
-import torch
+from pathlib import Path
 
+import torch
 from joblib import Parallel, delayed
-from pathlib import Path
 from sbi.inference import MNLE
 from sbi.utils import likelihood_nn
 
-
 BASE_DIR = Path(__file__).resolve().parent.parent.parent.as_posix()
 data_folder = BASE_DIR + "/data/"
 save_folder = BASE_DIR + "/notebooks/mnle-lan-comparison/models/"

notebooks/utils.py

@@ -1,21 +1,19 @@
 import math
 import os
-from pathlib import Path
 import pickle
+from pathlib import Path
 from typing import Any, Tuple
 
+import lanfactory
+import matplotlib.pyplot as plt
 import numpy as np
 import pandas as pd
-import matplotlib.pyplot as plt
 import torch
 from omegaconf import OmegaConf
+from sbi.inference.potentials.base_potential import BasePotential
 from sbibm.utils.io import get_float_from_csv
-from tqdm.auto import tqdm
-
 from torch.distributions.transforms import AffineTransform
-
-import lanfactory
-from sbi.inference.potentials.base_potential import BasePotential
+from tqdm.auto import tqdm
 
 
 def compile_df(