neural-amp-modeler

commit 044db9831892e83d4844c8313184fe9531b20f85
parent 9d6525e336461ac566995e12d6b9dabe538aaff2
Author: Steven Atkinson <steven@atkinson.mn>
Date:   Tue,  3 May 2022 17:45:41 -0700

Fiv env name, change plots, update README

Diffstat:
M
README.md
 | 
5
+++++
M
bin/train/main.py
 | 
9
++++++---
M
environment.yml
 | 
2
+-

3 files changed, 12 insertions(+), 4 deletions(-)
diff --git a/README.md b/README.md
@@ -6,6 +6,9 @@ For the code to create the plugin with a trained model, see my
 
 ## How to use
 
+This repository handles training, reamping, and exporting the weights of a model (to
+use with [the iPlug2 plugin]())
+
 ### Train a model
 
 You'll need at least two mono wav files: the input (DI) and the amped sound (without the cab).
@@ -33,6 +36,8 @@ bin/train/outputs/MyAmp
 Handy if you want to just check it out without going through the trouble of building the
 plugin.
 
+For example:
+
 ```bash
 python bin/run.py \
 path/to/source.wav \
diff --git a/bin/train/main.py b/bin/train/main.py
@@ -58,9 +58,12 @@ def plot(
         print(f"Took {t1 - t0} ({tx / (t1 - t0):.2f}x)")
 
     plt.figure(figsize=(16, 5))
-    plt.plot(ds.x[window_start:window_end], label="Input")
-    plt.plot(output[window_start:window_end], label="Output")
-    plt.plot(ds.y[window_start:window_end], label="Target")
+    # plt.plot(ds.x[window_start:window_end], label="Input")
+    plt.plot(output[window_start:window_end], label="Prediction")
+    plt.plot(ds.y[window_start:window_end], linestyle="--", label="Target")
+    # plt.plot(
+    #     ds.y[window_start:window_end] - output[window_start:window_end], label="Error"
+    # )
     plt.title(f"NRMSE={_rms(torch.Tensor(output) - ds.y) / _rms(ds.y)}")
     plt.legend()
     if savefig is not None:
diff --git a/environment.yml b/environment.yml
@@ -2,7 +2,7 @@
 # Created Date: Saturday February 13th 2021
 # Author: Steven Atkinson (steven@atkinson.mn)
 
-name: nam2
+name: nam
 channels:
   - pytorch
 dependencies: