CoolFace
Apppublic

NMR-CeNT-UW/ShimNet-development

sourceHugging Facemitupdated 6mo agoView on Hugging Face
0likes
App README

ShimNet

ShimNet is a data-driven AI solution to improve high-resolution nuclear magnetic resonance (NMR) spectra distorted by the inhomogeneous magnetic field (less than optimal shimming). To use it, the experimental training data has to be collected (see Data collection below). Example data can also be downloaded (see below).

Paper: ShimNet: A neural network for post-acquisition improvement of NMR spectra distorted by magnetic-field inhomogeneity

Web service: ![Open in Hugging Face Spaces](https://huggingface.co/spaces/NMR-CeNT-UW/ShimNet)

Installation

Python 3.10+ is required

You may install CPU-only version for inference only. If you need both training and inference, GPU version is strongly recommended.

In both CPU-only and GPU versions you may also install GUI (graphical user interface for inference)

  • —CPU-only version
  • —with GUI (recommended): pip install .[cpu,gui] --extra-index-url https://download.pytorch.org/whl/cpu
  • —without GUI: pip install .[cpu] --extra-index-url https://download.pytorch.org/whl/cpu
  • —GPU version (strongly recommended for training)
  • —with GUI: pip install .[gpu,gui]
  • —without GUI: pip install .[gpu]

Usage

To correct spectra presented in the paper:

  1. 1.download weights (model parameters):
python download_files.py

or directly from Google Drive 700MHz and Google Drive 600MHz and place it in weights directory

  1. 1.: run correction (e.g. Azarone_20ul_700MHz.csv):
python predict.py sample_data/Azarone_20ul_700MHz.csv -o output --config configs/shimnet_700.yaml --weights weights/shimnet_700MHz.pt

The output will be output/Azarone_20ul_700MHz_processed.csv file

Multiple files may be processed using "*" syntax:

python predict.py sample_data/*700MHz.csv -o output --config configs/shimnet_700.yaml --weights weights/shimnet_700MH
z.pt

For 600 MHz data use --config configs/shimnet_600.yaml and --weights weights/shimnet_600MHz.pt, e.g.:

python predict.py sample_data/CresolRed_after_styrene_600MHz.csv -o output --config configs/shimnet_600.yaml --weights weights/shimnet_600MHz.pt

input format

The spectrum file for reconstruction should be in the format of two columns separated by a space and without the sign at the end of the line at the end of the file. The first column is frequency in ppm, the second is the intensity. The frequency values ​​should be in ascending order (example below):

csv
-1.97134	0.0167137	
-1.97085	-0.00778748	
-1.97036	-0.0109595	
-1.96988	0.00825978	
-1.96939	0.0133886

Train on your data

For the model to function properly, it should be trained on calibration data from the spectrometer used for the measurements. To train a model on data from your spectrometer, please follow the instructions below.

Training data collection

Below we describe the training data collection for Agilent/Varian spectrometers. For machines of other vendors similar procedure can be implemented. To collect ShimNet training data use Python script (sweepshimslineshapeZ1Z2.py) from the calibrationloop folder to drive the spectrometer:

  1. 1.Install TReNDS package ( trends.spektrino.com )
  2. 2.Open VnmrJ and type: 'listenon'
  3. 3.Put the lineshape sample (1% CHCl3 in deuterated acetone), set standard PROTON parameters, and set nt=1 (do not modify sw and at!)
  4. 4.Shim the sample and collect the data. Save the optimally shimmed dataset
  5. 5.Edit the sweepshimslineshape_Z1Z2.py script
  6. 6.Put optimum z1 and z2 shim values as optiz1 and optiz2 below
  7. 7.Define the calibration range as rangez1 and rangez2 (default is ok)
  8. 8.Start the python script:
     python3 ./sweep_shims_lineshape_Z1Z2.py

The spectrometer will start collecting spectra

SCRF extraction

Shim Coil Response Functions (SCRF) should be extracted from the spectra with extract_scrf_from_fids.py script.

python extract_scrf_from_fids.py

The script uses hardcoded paths to the NMR signals (fid-s) in Agilent/Varian format: a directory with optimal measurement (opti_fid_path available) and a directory with calibration loop measurements (data_dir):

python
# input
data_dir = "../../sample_run/loop"
opti_fid_path = "../../sample_run/opti.fid"

The output files are also hardcoded:

python
# output
spectra_file = "../../sample_run/total.npy"
spectra_file_names = "../../sample_run/total.csv"
opi_spectrum_file = "../../sample_run/opti.npy"
responses_file = "../../sample_run/scrf_61.pt"

where only the responses_file is used in ShimNet training.

If the measurements are stored in a format other than Varian, you may need to change this line:

python
dic, data = ng.varian.read(varian_fid_path)

(see nmrglue package documentation for details)

Training

  1. 1.Download multiplets database:
    python download_files.py --multiplets
  1. 1.Configure run:
  2. 2.create a run directory, e.g. runs/my_lab_spectrometer_2025
  3. 3.create a configuration file:
  4. 4.copy configs/shimnet_template.py to the run directory and rename it to config.yaml
bash
       cp configs/shimnet_template.py runs/my_lab_spectrometer_2025/config.yaml
  1. 1.edit the SCRF in path in the config file:
yaml
         response_functions_files:
         - path/to/srcf_file

e.g.

yaml
         response_functions_files:
         - ../../sample_run/scrf_61.pt
  1. 1.adjust spectrometer frequency step frq_step to match your data (spectrometer range in Hz divided by number of points in spectrum):
yaml
        frq_step: 0.34059797
  1. 1.adjust spectromer frequency in the metadata
yaml
        metadata: # additional metadata, not used in the training process
          spectrometer_frequency: 700.0 # MHz
  1. 1.Run training:
    python train.py runs/my_lab_spectrometer_2025

Training results will appear in runs/my_lab_spectrometer_2025 directory. Model parameters are stored in runs/my_lab_spectrometer_2025/model.pt file

  1. 1.Use trained model:

use --config runs/my_lab_spectrometer_2025/config.yaml and --weights runs/my_lab_spectrometer_2025/model.pt flags, e.g.

    python predict.py my_sample1.csv -o my_output --config runs/my_lab_spectrometer_2025/config.yaml --weights runs/my_lab_spectrometer_2025/model.pt

Repeat training on our data

If you want to train the network using the calibration data from our paper, follow the procedure below.

  1. 1.Download multiplets database and our SCRF files:
    python download_files.py --multiplets --SCRF --no-weights

or directly download from Google Drive and store in data/ directory: Response Functions 600MHz, Response Functions 700MHz, Multiplets data

  1. 1.Configure run
  2. 2.For 600MHz spectrometer:
bash
      mkdir -p runs/repeat_paper_training_600MHz
      cp configs/shimnet_600.yaml runs/repeat_paper_training_600MHz/config.yaml
  • —For 700 MHz spectrometer:
bash
      mkdir -p runs/repeat_paper_training_700MHz
      cp configs/shimnet_700.yaml runs/repeat_paper_training_700MHz/config.yaml
  1. 1.Run training:
    python train.py runs/repeat_paper_training_600MHz

or

    python train.py runs/repeat_paper_training_700MHz

Training results will appear in runs/repeat_paper_training_600MHz or runs/repeat_paper_training_700MHz directory.

GUI

Launching the GUI

The ShimNet GUI is built using Gradio. To start the application, run:

bash
python predict-gui.py

Once the application starts, open your browser and navigate to:

http://127.0.0.1:7860

to access the GUI locally.

Sharing the GUI

To make the GUI accessible over the internet, use the --share flag:

bash
python predict-gui.py --share

A public web address will be displayed in the terminal, which you can use to access the GUI remotely or share with others.

GUI inference with Docker

Create docker image:

bash
docker build -t shimnetgui .

Run the container:

bash
docker run -it -p 7860:7860 shimnetgui

The GUI should be working at http://127.0.0.1:7860