Let twin-axis aligned at the specified position #26109
Conversation
| 0, 0, 30, 0, 0, 0, 0, 0, -30, 0] | ||
| net_gain= [0, -2, -3, 5, 2, 3, 4, 5, 5, -1, | ||
| -4, -10, 2, 5, 9, 6, 0, 1, -1, 9] | ||
| df = pd.DataFrame({"net_in": net_in, "net_gain": net_gain}) |
There was a problem hiding this comment.
Can this be done in numpy instead of pandas?
There was a problem hiding this comment.
Yes, I have already done this with numpy.
| df["total_in"] = df["net_in"].cumsum() | ||
| df["value"] = df["total_in"] + df["net_gain"].cumsum() | ||
| df["value_net"] = df["value"] / df["value"].iloc[0] | ||
| df["gain_pct"] = df["value"] / df["total_in"] - 1 |
There was a problem hiding this comment.
I don't think you need all these derived variables?
There was a problem hiding this comment.
Yes, I have omitted the data generation process and provided the final data that needs to be plotted directly.
| plt.show() | ||
|
|
||
|
|
||
| general_plot() |
There was a problem hiding this comment.
I would just in-line these. If you want to show both, I'd use two axes
fig, axs = plt.subplots(2, 1, layout='constrained')
for nn in range(2):
ax1 = axs[nn]
...plotting
if nn == 1:
twinxalign(ax1, ax2, 1, 0)
ax.title('align')
else:
ax.title('not aligned')
There was a problem hiding this comment.
I have merged the drawing process into one function and used the parameter align to control whether to align or not.
| def _inverse(x): | ||
| return (x - b_new) / k_new | ||
| ax_right.set_ylim([right_min_new, right_max_new]) | ||
| ax_right.set_yscale("function", functions=(_forward, _inverse)) |
There was a problem hiding this comment.
I'm actually not sure why you want a scale here. It's just linear between min/max_new?
There was a problem hiding this comment.
My original intention was to confirm the linear map of right_min_new-right_max_new and right_min-right_max is performed.
But I found this step seems be already done in the set_ylim step, so there is no need to do it again. I have removed it in the new code.
PR summary
Let the left-axis and right-axis aligned at the specified position.
In some data that need twin-axis plot, a point at left-axis and another point at right-axis have same meaning, they shoud be aligned.
For example, we plot netvalue curve of a portfolio at twin-left-axis and profit raito curve at twin-right-axis, the point 1.0 at the
left-axis and the point 0.0 at the right-axis both mean the begin state of the portifolio, so they should be aligned.
general twinx plot:
twinx aligned plot:
PR checklist
[✔] new and changed code is tested