This is a version of https://github.com/sdatkinson/neural-amp-modeler that has a batch script so you can queue training.

Using the batch script is fairly simple, but there are some setup rules you will need to follow.

Installing is easy, first follow the install instruction for the original neural-amp-modeler if you haven't already, then install my custom repo that contains the batch process.

briefly: Install Conda https://docs.conda.io/projects/conda/en/latest/user-guide/install/index.html using conda, install git if you don't have it already

conda install git

git clone https://github.com/sdatkinson/neural-amp-modeler

cd neural-amp-modeler

conda env create -f environment_gpu.yml

python setup.py install

cd ..

git clone https://github.com/rossbalch/nam-batch

Usage

Copy your input and output files to the materials folder in the nam-batch directory. It is important that you copy the files and not move them, the files will be deleted after the process finishes. There are also some naming considerations. Your files must be called name_In.wav and name_Out.wav the batch script will not recognise them other wise.

For Instance:
Peavy_5150_In.wav
Peavy_5150_Out.wav

If you use the original v_1_1_1 file just copy and rename it for each model you wish to train.
Once all the files you wish to train are in the materials folder simply run the script, it will ascertain the delay, queue up the training, generate the models and all the resulting files will be available in the exports folder.

Please note at the moment the quoted ESRs are actually MRSE, so to get the ESR you need to perform the quoted number^2

cd nam-batch

conda activate nam

python batch.py

Troubleshooting

If you get into some issues, for instance the ESR is bad, it could be the auto determined delay value is bad, in this case it might be worth running the batching in 3 steps instead. prep_batch.py will create all of the folders and config files neccessary to run in a batch, at this point you can manually change the delay value in delay.csv Then you can run train_batch.py and finally export_batch.py . If the batch process runs into error clean.py will put the batch directory back to the creation state, you will need to re-copy your input and output files.