File size: 3,338 Bytes
5a5d1a8 | 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 | import sys
from pathlib import Path
# 获取项目根目录（inference.py上级的上级）
root_path = Path(__file__).parent.parent
sys.path.append(str(root_path))
import torch
import os
import glob
import numpy as np
import h5py
from tqdm import tqdm
from model.dgmr import DGMR
from onescience.utils.YParams import YParams
from onescience.datapipes.climate import ERA5Datapipe
def get_stats(data_dir, channels):
"""从新版 h5 中读取变量列表与归一化参数（均值/标准差）"""
h5_files = sorted(glob.glob(os.path.join(data_dir, "data", "*.h5")))
with h5py.File(h5_files[0], "r") as f:
ds = f["fields"]
all_variables = [v.decode() if isinstance(v, bytes) else v for v in ds.attrs["variables"]]
mu = f["global_means"][:] # [1, C, 1, 1]
std = f["global_stds"][:]
channel_indices = [all_variables.index(v) for v in channels]
means = mu[:, channel_indices, :, :]
stds = std[:, channel_indices, :, :]
return means, stds
if __name__ == "__main__":
current_path = os.getcwd()
sys.path.append(current_path)
## Model config init
config_file_path = os.path.join(current_path, "conf/config.yaml")
cfg = YParams(config_file_path, "model")
## DataLoader init
cfg_data = YParams(config_file_path, "datapipe")
means, stds = get_stats(cfg_data.dataset.data_dir, cfg_data.dataset.channels)
cfg['N_in_channels'] = len(cfg_data.dataset.channels)
cfg['N_out_channels'] = len(cfg_data.dataset.channels)
datapipe = ERA5Datapipe(
dataset_dir=cfg_data.dataset.data_dir,
used_variables=cfg_data.dataset.channels,
used_years=cfg_data.dataset.test_time,
distributed=False,
input_steps=cfg.num_context,
output_steps=cfg.forecast_steps,
batch_size=1,
num_workers=4,
)
test_dataloader, _ = datapipe.get_dataloader("test")
device = "cuda:0" if torch.cuda.is_available() else "cpu"
ckpt = torch.load(f"{cfg.checkpoint_dir}/model_bak.pth", map_location=device, weights_only=False)
model = DGMR(
forecast_steps=cfg.forecast_steps,
num_context=cfg.num_context,
input_channels=cfg.input_channels,
output_shape=cfg.output_shape,
conv_type=cfg.conv_type,
latent_channels=cfg.latent_channels,
context_channels=cfg.context_channels,
generation_steps=cfg.generation_steps,
grid_lambda=cfg.grid_lambda,
precip_weight_cap=cfg.precip_weight_cap,
).to(device)
model.load_state_dict(ckpt["model_state_dict"])
model.eval()
os.makedirs('result/output/', exist_ok=True)
print(f"📂 infer results will be generated to './result/output/'")
with torch.no_grad():
for data in tqdm(test_dataloader, desc="Inferring testset", unit="batch"):
invar = data[0].to(device, dtype=torch.float32) # [1, num_context, C, H, W]
gen = model.generator(invar).cpu().numpy() # [1, forecast_steps, C, H, W]
for t in range(gen.shape[1]):
fname = data[4][cfg.num_context + t][0] # 该预测帧对应的时刻
pred_var = gen[0, t] # [C, H, W]
pred_var = pred_var * stds + means
np.save(f"result/output/{fname}.npy", pred_var)
|