From 99ba4e390ea4a446f359da611a92761d73db0e7f Mon Sep 17 00:00:00 2001
From: "matthias@arch" <matthiasqui@protonmail.com>
Date: Wed, 31 May 2023 11:04:37 +0200
Subject: [PATCH] Fixed SimpleNormalize, added NormalizeAmplitude

---
 teng_ml/util/transform.py | 41 +++++++++++++++++++++++++++------------
 1 file changed, 29 insertions(+), 12 deletions(-)

diff --git a/teng_ml/util/transform.py b/teng_ml/util/transform.py
index 96b7f5b..db78cc6 100644
--- a/teng_ml/util/transform.py
+++ b/teng_ml/util/transform.py
@@ -9,20 +9,37 @@ class Normalize:
         assert(low < high)
         self.low = low
         self.high = high
-    def __call__(self, a):
-        min_ = np.min(a)
-        a = a - min_
-        max_ = np.max(a)
+    def __call__(self, data):
+        min_ = np.min(data)
+        data = data - min_   # smallest point is 0 now
+        max_ = np.max(data)
         if max_ != 0:
-            a = (a / max_)
+            data = (data / max_)
         # now normalized between 0 and 1
-        a *= (self.high - self.low)
-        a -= self.low
-        return a
+        data *= (self.high - self.low)
+        data += self.low
+        return data
 
     def __repr__(self):
         return f"Normalize(low={self.low}, high={self.high})"
 
+class NormalizeAmplitude:
+    """
+    scale data so that all values are between -high and high
+    """
+    def __init__(self, high=1):
+        self.high = high
+
+    def __call__(self, data):
+        min_ = np.min(data)
+        max_ = np.max(data)
+        scale = np.max([np.abs(min_), np.abs(max_)])
+        if scale != 0:
+            data = data / scale * self.high
+        return data
+    def __repr__(self):
+        return f"NormalizeAmplitude(high={self.high})"
+
 
 class ConstantInterval:
     """
@@ -32,15 +49,15 @@ class ConstantInterval:
     def __init__(self, interval):
         self.interval = interval
 
-    def __call__(self, a):
+    def __call__(self, data):
         """
         array: [timestamps, data1, data2...]
         """
-        timestamps = a[:,0]
+        timestamps = data[:,0]
         new_stamps = np.arange(timestamps[0], timestamps[-1], self.interval)
         ret = new_stamps
-        for i in range(1, a.shape[1]):  # 
-            interp = interp1d(timestamps, a[:,i])
+        for i in range(1, data.shape[1]):  # 
+            interp = interp1d(timestamps, data[:,i])
             new_vals = interp(new_stamps)
             ret = np.vstack((ret, new_vals))
         return ret.T