nnUNetv2_plan_and_preprocess: error: unrecognized arguments: --npfp 8 Exit code: 2 Preprocessing directory not created

[Satakang483BIMIT]

NNU_Net_12_12_24 (2).md

Good morning, I am currently trying to create a detailed and well annotated script for training using nn-U-NETv2. The script is going include visualization of key metrics during training and give me key visual graphs/charts for model evaluation afterwards.

My goal is to train the model using labels already created from manually segmenting MRIs in the MONAI Label module of 3DSlicer.

Right now, I am running the script in Google Colab via Local Host.

Everything in the script runs as expected until this current block of code:

"# Modified preprocessing script with basic arguments
import os
import subprocess

cmd = "nnUNetv2_plan_and_preprocess -d 100 --verify_dataset_integrity --npfp 8"

try:

Run preprocessing with full output capture

result = subprocess.run(
cmd,
shell=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True
)

print("Command Output:")
print(result.stdout)

print("\nError Output (if any):")
print(result.stderr)

Print exit code

print("\nExit code:", result.returncode)

Check if preprocessing succeeded

preprocessed_path = os.path.join(os.environ.get('nnUNet_preprocessed'), 'Dataset100_CSFSegmentation')
if os.path.exists(preprocessed_path):
print(f"\nPreprocessing completed successfully")
print("Contents:", os.listdir(preprocessed_path))
else:
print("\nPreprocessing directory not created")

except Exception as e:
print(f"Error running preprocessing: {e}")"

Resulting Output:

"Command Output:

Error Output (if any):
usage: nnUNetv2_plan_and_preprocess [-h] [-d D [D ...]] [-fpe FPE]
[-npfp NPFP] [--verify_dataset_integrity]
[--no_pp] [--clean] [-pl PL]
[-gpu_memory_target GPU_MEMORY_TARGET]
[-preprocessor_name PREPROCESSOR_NAME]
[-overwrite_target_spacing OVERWRITE_TARGET_SPACING [OVERWRITE_TARGET_SPACING ...]]
[-overwrite_plans_name OVERWRITE_PLANS_NAME]
[-c C [C ...]] [-np NP [NP ...]]
[--verbose]
nnUNetv2_plan_and_preprocess: error: unrecognized arguments: --npfp 8

Exit code: 2

Preprocessing directory not created"

I have attempted several troubleshooting fixes (in both the code itself & bash/Anaconda Powershell Prompt), such as:

1. "nnUNetv2_plan_and_preprocess -d 100 --verify_dataset_integrity --clean -npfp 8"

RESULT: : AssertionError: Did not find the expected number of training cases (90). Found 1 instead.
Examples: ['CS']

2. "nnUNetv2_plan_and_preprocess -d 100 --verify_dataset_integrity --clean"

RESULT: AssertionError: Did not find the expected number of training cases (90). Found 1 instead.
Examples: ['CS']

3. "nnUNetv2_plan_and_preprocess -d 100 --verify_dataset_integrity --clean --np 8"
   RESULT: nnUNetv2_plan_and_preprocess: error: unrecognized arguments: --np 8

What further troubleshooting steps could I take to establish the preprocessing directory?

[Satakang483BIMIT]

Here is the result of the Current Google Colab code

[Satakang483BIMIT]

dataset.json
Attached is the .json file in the "C:\Users\Administrator\Desktop\ColabNotebooks\nnUNet_raw\Dataset100" directory

Here are screengrabs of the dataset.json file: