ModuleNotFoundError: No module named "'step" about step HOT 2 CLOSED

Please make sure to run the commands in the README in the directory C:\path\to\STEP, instead of C:\path\to\STEP\step.
Reference: Issue #17

from step.

Hi zezhishao，
Thanks for your kind reply, I found that this problem may be caused by my device. When I run the jupyter notebook in the C: path to STEP directory:! python step/run. py --cfg='step/STEP_ METR-LA. After py '-- gpus='0', you will get the same error as shown in the following ：

2022-12-06 09:11:24,320 - easytorch-launcher - INFO - Launching EasyTorch training. Traceback (most recent call last): File "D:\python-venv\STEP\step\run.py", line 26, in <module> launch_training(args.cfg, args.gpus) File "D:\python-venv\STEP\basicts\launcher.py", line 19, in launch_training easytorch.launch_training(cfg=cfg, gpus=gpus, node_rank=node_rank) File "D:\python-venv\venv\lib\site-packages\easytorch\launcher\launcher.py", line 80, in launch_training cfg = init_cfg(cfg, node_rank == 0) File "D:\python-venv\venv\lib\site-packages\easytorch\launcher\launcher.py", line 13, in init_cfg cfg = import_config(cfg, verbose=save) File "D:\python-venv\venv\lib\site-packages\easytorch\config.py", line 245, in import_config cfg = __import__(path, fromlist=[cfg_name]).CFG ModuleNotFoundError: No module named "'step"

When running under the terminal, the expected effect is obtained：

`2022-12-06 15:45:45,971 - easytorch-launcher - INFO - Launching EasyTorch training.
DESCRIPTION: STEP(METR-LA) configuration
RUNNER: <class 'step.step_runner.step_runner.STEPRunner'>
DATASET_CLS: <class 'step.step_data.forecasting_dataset.ForecastingDataset'>
DATASET_NAME: METR-LA
DATASET_TYPE: Traffic speed
DATASET_INPUT_LEN: 12
DATASET_OUTPUT_LEN: 12
DATASET_ARGS:
seq_len: 2016
GPU_NUM: 1
ENV:
SEED: 0
CUDNN:
ENABLED: True
MODEL:
NAME: STEP
ARCH: <class 'step.step_arch.step.STEP'>
PARAM:
dataset_name: METR-LA
pre_trained_tsformer_path: tsformer_ckpt/TSFormer_METR-LA.pt
tsformer_args:
patch_size: 12
in_channel: 1
embed_dim: 96
num_heads: 4
mlp_ratio: 4
dropout: 0.1
num_token: 168.0
mask_ratio: 0.75
encoder_depth: 4
decoder_depth: 1
mode: forecasting
backend_args:
num_nodes: 207
supports: [tensor([[0.2050, 0.0000, 0.0000, ..., 0.0000, 0.0000, 0.0000],
[0.0000, 0.2626, 0.1503, ..., 0.0000, 0.0000, 0.0000],
[0.0000, 0.1027, 0.2095, ..., 0.0000, 0.0000, 0.0000],
...,
[0.0000, 0.0000, 0.0000, ..., 0.2788, 0.0000, 0.0000],
[0.0000, 0.0000, 0.0000, ..., 0.0000, 0.2645, 0.0000],
[0.0000, 0.0000, 0.0000, ..., 0.0000, 0.0000, 0.1408]]), tensor([[0.2452, 0.0000, 0.0000, ..., 0.0000, 0.0000, 0.0000],
[0.0000, 0.1789, 0.0968, ..., 0.0000, 0.0000, 0.0000],
[0.0000, 0.1283, 0.2475, ..., 0.0000, 0.0000, 0.0000],
...,
[0.0000, 0.0000, 0.0000, ..., 0.4463, 0.0000, 0.0000],
[0.0000, 0.0000, 0.0000, ..., 0.0000, 0.2833, 0.0000],
[0.0000, 0.0000, 0.0000, ..., 0.0000, 0.0000, 0.1831]])]
dropout: 0.3
gcn_bool: True
addaptadj: True
aptinit: None
in_dim: 2
out_dim: 12
residual_channels: 32
dilation_channels: 32
skip_channels: 256
end_channels: 512
kernel_size: 2
blocks: 4
layers: 2
dgl_args:
dataset_name: METR-LA
k: 10
input_seq_len: 12
output_seq_len: 12
FROWARD_FEATURES: [0, 1, 2]
TARGET_FEATURES: [0]
DDP_FIND_UNUSED_PARAMETERS: True
TRAIN:
LOSS: step_loss
OPTIM:
TYPE: Adam
PARAM:
lr: 0.005
weight_decay: 1e-05
eps: 1e-08
LR_SCHEDULER:
TYPE: MultiStepLR
PARAM:
milestones: [1, 18, 36, 54, 72]
gamma: 0.5
CLIP_GRAD_PARAM:
max_norm: 3.0
NUM_EPOCHS: 100
CKPT_SAVE_DIR: checkpoints\STEP_100
DATA:
DIR: datasets/METR-LA
BATCH_SIZE: 8
PREFETCH: False
SHUFFLE: True
NUM_WORKERS: 2
PIN_MEMORY: True
NULL_VAL: 0.0
CL:
WARM_EPOCHS: 0
CL_EPOCHS: 6
PREDICTION_LENGTH: 12
VAL:
INTERVAL: 1
DATA:
DIR: datasets/METR-LA
BATCH_SIZE: 8
PREFETCH: False
SHUFFLE: False
NUM_WORKERS: 2
PIN_MEMORY: True
TEST:
INTERVAL: 1
DATA:
DIR: datasets/METR-LA
BATCH_SIZE: 8
PREFETCH: False
SHUFFLE: False
NUM_WORKERS: 2
PIN_MEMORY: True

2022-12-06 15:45:46,423 - easytorch-env - INFO - Use GPUs 0.
2022-12-06 15:45:46,455 - easytorch-launcher - INFO - Initializing runner "<class 'step.step_runner.step_runner.STEPRunner'>"
2022-12-06 15:45:46,455 - easytorch-env - INFO - Disable TF32 mode
2022-12-06 15:45:46,455 - easytorch - INFO - Set ckpt save dir: 'checkpoints\STEP_100\53a27f003182ec852ca7bad1e6ca286a'
2022-12-06 15:45:46,455 - easytorch - INFO - Building model.
2022-12-06 15:45:47,139 - easytorch-training - INFO - Initializing training.
2022-12-06 15:45:47,139 - easytorch-training - INFO - Set clip grad, param: {'max_norm': 3.0}
2022-12-06 15:45:47,139 - easytorch-training - INFO - Building training data loader.
train len: 23974
2022-12-06 15:45:47,291 - easytorch-training - INFO - Set optim: Adam (
Parameter Group 0
amsgrad: False
betas: (0.9, 0.999)
capturable: False
differentiable: False
eps: 1e-08
foreach: None
fused: False
lr: 0.005
maximize: False
weight_decay: 1e-05
)
2022-12-06 15:45:47,292 - easytorch-training - INFO - Set lr_scheduler: <torch.optim.lr_scheduler.MultiStepLR object at 0x000001774A382FA0>
2022-12-06 15:45:47,299 - easytorch-training - INFO - Initializing validation.
2022-12-06 15:45:47,300 - easytorch-training - INFO - Building val data loader.
val len: 3425
test len: 6850
2022-12-06 15:45:47,496 - easytorch-training - INFO - Epoch 1 / 100
0%| `

from step.