har_trees: Initial script for reading data

The values look quite suspicious right now

Author: jonnor

examples/har_trees/read_data.py

@@ -0,0 +1,47 @@
+
+
+import os
+
+import pandas
+import numpy
+
+def load_har_record(path, samplerate):
+    suffix = '.npy'
+
+    files = []
+
+    for f in os.listdir(path):
+        if f.endswith(suffix):
+            p = os.path.join(path, f)
+            try:
+                data = numpy.load(p, allow_pickle=True)
+            except Exception as e:
+                print(e)
+                continue
+
+            df = pandas.DataFrame(data.T, columns=['x', 'y', 'z'])
+            t = numpy.arange(0, len(df)) * (1.0/samplerate)
+            df['time'] = t
+            df = df.set_index('time')
+            classname = f.split('_')[1].rstrip(suffix)
+
+            files.append(dict(data=df, filename=f, classname=classname))
+
+            #print(f, data.shape)
+
+    out = pandas.DataFrame.from_records(files)
+    out = out.set_index('filename')
+    return out
+
+p = '/home/jon/temp/micropython-test-esptool/har_record'
+data = load_har_record(p, samplerate=100)
+
+print(data.head())
+print(data.shape)
+
+import plotly.express
+
+for idx, f in data.iterrows():
+    d = f.data.reset_index()
+    fig = plotly.express.line(d, x='time', y=['x', 'y', 'z'])
+    fig.show()