Tidy up these figure generators

ch-13/types_of_randomness/msgpack_demo.py

@@ -1,31 +0,0 @@
-from ulab import numpy as np
-import random
-import msgpack
-from io import BytesIO
-
-data = np.array([[random.uniform(-200, 200), random.uniform(-200, 200), random.uniform(0, 360)] for i in range(20)], dtype=np.int16)
-print(data)
-print(data.shape)
-buffer = BytesIO()
-msgpack.pack({"offset": 10, "poses": data.tolist()},  buffer)
-buffer.seek(0)
-print(len(buffer.getvalue())) # json is 618 bytes, msgpack is 595 bytes. msgpack is 3.7% smaller
-print(buffer.getvalue())
-
-import json
-as_int = np.array(data, dtype=np.int16)
-as_int_json = json.dumps({"offset": 10, "poses": as_int.tolist()})
-
-
-## on pc
-import msgpack
-import numpy as np
-raw_data = "bytes from robot"
-data = msgpack.unpackb(raw_data)
-poses = np.array(data["poses"])
-print(poses)
-# Avoiding the text could make it smaller.
-# From https://learn.adafruit.com/introducing-the-adafruit-bluefruit-le-uart-friend/hardware
-# Note that we do not recommend using higher baudrates than 9600 because the nRF51 UART can drop characters!
-# Is the complexity of the msgpack worth the 3.7% size reduction?
-

ch-13/types_of_randomness/pdfs_of_uniform_sums.py

@@ -7,8 +7,13 @@ population_size = 100000
 
 def make_uniform_series_plot(n):
     uniform_series = np.array(
-        [sum(random.uniform(0, 1) for _ in range(n))/n for _ in range(population_size)])
-    plt.hist(uniform_series, bins=200, histtype='step', label=f"n={n}")
+        [
+            sum(random.uniform(0, 1) for _ in range(n)) / n
+            for _ in range(population_size)
+        ]
+    )
+    plt.hist(uniform_series, bins=200, histtype="step", label=f"n={n}")
+
 
 make_uniform_series_plot(1)
 make_uniform_series_plot(2)

ch-13/types_of_randomness/show_random_distributions.py

@@ -5,6 +5,7 @@ distribution_size = (2, 500)
 uniform_dist = np.random.uniform(low=-1.5, high=1.5, size=distribution_size)
 gauss_dist = np.random.normal(loc=0.0, scale=0.5, size=distribution_size)
 
+
 def prepare_scatter(ax, data):
     ax.set_xlim(-1.5, 1.5)
     ax.set_ylim(-1.5, 1.5)
@@ -13,21 +14,22 @@ def prepare_scatter(ax, data):
     data_sd = data.std(1)
     data_mean = data.mean(1)
     print(data_mean, data_sd)
-    circle = plt.Circle(data_mean, data_sd[0], color='r', fc=(1,0.6,0.6,0.3))
+    circle = plt.Circle(data_mean, data_sd[0], color="r", fc=(1, 0.6, 0.6, 0.3))
     ax.add_patch(circle)
 
+
 fig = plt.figure()
 gs = fig.add_gridspec(2, 2)
 uniform_scatter = fig.add_subplot(gs[0, 0])
 gauss_scatter = fig.add_subplot(gs[0, 1])
-uniform_hist = fig.add_subplot(gs[1, 0], sharex = uniform_scatter)
-gauss_hist = fig.add_subplot(gs[1, 1], sharex = gauss_scatter)
+uniform_hist = fig.add_subplot(gs[1, 0], sharex=uniform_scatter)
+gauss_hist = fig.add_subplot(gs[1, 1], sharex=gauss_scatter)
 # fig, ((uniform_scatter, gauss_scatter), (uniform_hist, gauss_hist)) = plt.subplots(2, 2)
 prepare_scatter(uniform_scatter, uniform_dist)
 prepare_scatter(gauss_scatter, gauss_dist)
 
 # Plot the histograms
-bins=16
+bins = 16
 uniform_hist.hist(uniform_dist[0], bins=bins)
 gauss_hist.hist(gauss_dist[0], bins=bins)
 # fig.tight_layout()