I wanna plot a scatter to visualize my result using matplotlib:
plt.subplot(1, 2, 1)
plt.scatter(source_weights, target_weights)
plt.xlabel('Source Weights', fontsize=13, fontweight='bold')
plt.ylabel('Target Weights', fontsize=13, fontweight='bold')
plt.xticks(fontsize=12, fontweight='bold')
plt.yticks(fontsize=12, fontweight='bold')
plt.axis('equal')
plt.axis('square')
y_lim = np.max(np.abs(target_weights))
x_lim = np.max(np.abs(source_weights))
lim = max(x_lim, y_lim)
_ = plt.plot([-1.1 * lim, 1.1 * lim], [-1.1 * lim, 1.1 * lim])
# plot bias difference
plt.subplot(1, 2, 2)
plt.scatter(source_bias, target_bias)
plt.xlabel('Source Bias', fontsize=13, fontweight='bold')
plt.ylabel('Target Bias', fontsize=13, fontweight='bold')
plt.xticks(fontsize=12, fontweight='bold')
plt.yticks(fontsize=12, fontweight='bold')
plt.axis('equal')
plt.axis('square')
y_lim = np.max(np.abs(target_bias))
x_lim = np.max(np.abs(source_bias))
lim = max(x_lim, y_lim)
_ = plt.plot([-1.1 * lim, 1.1 * lim], [-1.1 * lim, 1.1 * lim])
But I found that sometimes the scalars of the axis are stacked together, like this in the source bias:
Is there any method that can solve the problem without changing the font size, like setting fewer scale marks at the axis. I still want the same scaling of x- and y-axis (1:1 square) by the way.
