File size: 3,823 Bytes
ebbf522 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 | import numpy as np
import torch
from torch.utils.data import Dataset
import pickle
import h5py
import json
import linecache
output_path = '/data/beegfs/darcy/sionna_simulation/BigSim/'
def get_args(config_file, task_id):
# linecache is 1-indexed, which matches SLURM_ARRAY_TASK_ID perfectly
line = linecache.getline(config_file, task_id)
return json.loads(line)
class SICDFDataset(Dataset):
"""
"""
def __init__(self, task_id, mode = "train", angle_mode = "deg"):
args = get_args("config.jsonl", task_id)
print(f"Starting run with task_id {task_id} and arguments {args}...")
jammer_sig_file = args['jammer_file']
jammer_file = jammer_sig_file.split('/')[-1][:-4]
trajectory = args['trajectory']
ant_dist = args['ant_dist']
antenna_gain = args['antenna_gain']
layout = args['layout']
snr = args['snr']
reflection_depth = args['reflection_depth']
refraction = args['refraction']
data_file_name = output_path + f"{jammer_file}_{trajectory}_{ant_dist}_{antenna_gain}_{layout}_{snr}_{reflection_depth}_{refraction}"
self.angle_mode = angle_mode
self.h5_file = None
PATH_IQ_TRAIN = data_file_name + "_train.h5"
PATH_IQ_TEST = data_file_name + "_test.h5"
DATA_FILE_TRAIN = data_file_name + "_train.txt"
DATA_FILE_TEST = data_file_name + "_test.txt"
if mode == "train":
with open(DATA_FILE_TRAIN, "r") as f:
data_info = f.readlines()
elif mode == "test":
with open(DATA_FILE_TEST, "r") as f:
data_info = f.readlines()
else:
raise ValueError("Mode must be 'train', 'test', 'outside' or 'shifted'")
data_info = np.array([x.split("\n")[0].split("\t") for x in data_info])
data_info = data_info[1:]
if mode == "train":
if self.h5_file is None:
self.h5_file = h5py.File(PATH_IQ_TRAIN, 'r')
elif mode == "test":
if self.h5_file is None:
self.h5_file = h5py.File(PATH_IQ_TEST, 'r')
else:
raise ValueError("Mode must be 'train', 'test', 'outside' or 'shifted'")
self.x_files = []
self.positions = []
self.az_angles = []
self.el_angles = []
for x in data_info:
self.positions.append(np.array([float(x[1]), float(x[2]), float(x[3])]))
if self.angle_mode == "deg":
self.az_angles.append(float(x[4]))
self.el_angles.append(float(x[5]))
elif self.angle_mode == "rad":
self.az_angles.append(np.deg2rad(float(x[4])))
self.el_angles.append(np.deg2rad(float(x[5])))
elif self.angle_mode == "sincos":
self.az_angles.append([np.sin(np.deg2rad(float(x[4]))), np.cos(np.deg2rad(float(x[4])))])
self.el_angles.append([np.sin(np.deg2rad(float(x[5]))), np.cos(np.deg2rad(float(x[5])))])
else:
raise ValueError("angle_mode must be 'deg', 'rad' or 'sincos'")
self.x_files.append(float(x[0]))
self.az_angles = np.array(self.az_angles)
self.el_angles = np.array(self.el_angles)
self.positions = np.array(self.positions)
self.x_files = np.array(self.x_files)
def __getitem__(self, item: int):
az_angle = torch.tensor(self.az_angles[item], dtype=torch.float)
el_angle = torch.tensor(self.el_angles[item], dtype=torch.float)
position = torch.tensor(self.positions[item])
index = int(self.x_files[item])
return torch.tensor(self.h5_file["received_signals"][index], dtype=torch.complex64), (position, az_angle, el_angle)
def __len__(self) -> int:
return len(self.az_angles) |