17934599025

Committed 23 Sep 2025 03:07AM UTC coverage: 86.608% (+2.6%) from 83.977%

Build # 17934599025

Build Type

push

github

Committed by

avivajpeyi

Commit Message

remove sparcity comments

Run Details

274 of 308 branches covered (88.96%)

Branch coverage included in aggregate %.

2274 of 2634 relevant lines covered (86.33%)

1.73 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

95.12

/src/log_psplines/plotting/diagnostics.py

import arviz as az
import matplotlib.pyplot as plt
import numpy as np


def plot_diagnostics(
    idata: az.InferenceData,
    outdir: str,
    # variables: list = ["phi", "delta", "weights"],
    figsize: tuple = (12, 8),
) -> None:
    """
    Plot MCMC diagnostics using arviz.

    Parameters
    ----------
    idata : az.InferenceData
        Inference data from adaptive MCMC
    variables : list
        Variables to plot
    figsize : tuple
        Figure size
    """

    try:

        # Trace plots
        az.plot_trace(idata, figsize=figsize) #, var_names=variables)
        plt.suptitle("Trace plots - Adaptive MCMC")
        plt.tight_layout()
        plt.savefig(f"{outdir}/trace_plots.png")

        # Acceptance rate plot
        if "acceptance_rate" in idata.sample_stats:
            fig, ax = plt.subplots(figsize=(10, 4))
            accept_rates = idata.sample_stats.acceptance_rate.values.flatten()
            ax.plot(accept_rates, alpha=0.7)
            ax.axhline(
                idata.attrs.get("target_accept_rate", 0.44),
                color="red",
                linestyle="--",
                label="Target",
            )
            ax.set_xlabel("Iteration")
            ax.set_ylabel("Acceptance Rate")
            ax.set_title("Acceptance Rate Over Time")
            ax.legend()
            ax.grid(True, alpha=0.3)
            plt.tight_layout()
            plt.savefig(f"{outdir}/acceptance_rate.png")

        # Step size evolution
        if "step_size_mean" in idata.sample_stats:
            fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))

            step_means = idata.sample_stats.step_size_mean.values.flatten()
            step_stds = idata.sample_stats.step_size_std.values.flatten()

            ax1.plot(step_means, alpha=0.7)
            ax1.set_xlabel("Iteration")
            ax1.set_ylabel("Mean Step Size")
            ax1.set_title("Step Size Evolution")
            ax1.grid(True, alpha=0.3)

            ax2.plot(step_stds, alpha=0.7, color="orange")
            ax2.set_xlabel("Iteration")
            ax2.set_ylabel("Step Size Std")
            ax2.set_title("Step Size Variability")
            ax2.grid(True, alpha=0.3)

            plt.tight_layout()
            plt.savefig(f"{outdir}/step_size_evolution.png")

    except Exception as e:
        print(f"Error generating adaptive MCMC diagnostics: {e}")


    # NUTS-specific plots
    ## ... unsure what diagnostics we should add...

1	import arviz as az	2✔
2	import matplotlib.pyplot as plt	2✔
3	import numpy as np	2✔
4
5
6	def plot_diagnostics(	2✔
7	idata: az.InferenceData,
8	outdir: str,
9	# variables: list = ["phi", "delta", "weights"],
10	figsize: tuple = (12, 8),
11	) -> None:
12	"""
13	Plot MCMC diagnostics using arviz.
14
15	Parameters
16	----------
17	idata : az.InferenceData
18	Inference data from adaptive MCMC
19	variables : list
20	Variables to plot
21	figsize : tuple
22	Figure size
23	"""
24
25	try:	2✔
26
27	# Trace plots
28	az.plot_trace(idata, figsize=figsize) #, var_names=variables)	2✔
29	plt.suptitle("Trace plots - Adaptive MCMC")	2✔
30	plt.tight_layout()	2✔
31	plt.savefig(f"{outdir}/trace_plots.png")	2✔
32
33	# Acceptance rate plot
34	if "acceptance_rate" in idata.sample_stats:	2✔
35	fig, ax = plt.subplots(figsize=(10, 4))	2✔
36	accept_rates = idata.sample_stats.acceptance_rate.values.flatten()	2✔
37	ax.plot(accept_rates, alpha=0.7)	2✔
38	ax.axhline(	2✔
39	idata.attrs.get("target_accept_rate", 0.44),
40	color="red",
41	linestyle="--",
42	label="Target",
43	)
44	ax.set_xlabel("Iteration")	2✔
45	ax.set_ylabel("Acceptance Rate")	2✔
46	ax.set_title("Acceptance Rate Over Time")	2✔
47	ax.legend()	2✔
48	ax.grid(True, alpha=0.3)	2✔
49	plt.tight_layout()	2✔
50	plt.savefig(f"{outdir}/acceptance_rate.png")	2✔
51
52	# Step size evolution
53	if "step_size_mean" in idata.sample_stats:	2✔
54	fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))	2✔
55
56	step_means = idata.sample_stats.step_size_mean.values.flatten()	2✔
57	step_stds = idata.sample_stats.step_size_std.values.flatten()	2✔
58
59	ax1.plot(step_means, alpha=0.7)	2✔
60	ax1.set_xlabel("Iteration")	2✔
61	ax1.set_ylabel("Mean Step Size")	2✔
62	ax1.set_title("Step Size Evolution")	2✔
63	ax1.grid(True, alpha=0.3)	2✔
64
65	ax2.plot(step_stds, alpha=0.7, color="orange")	2✔
66	ax2.set_xlabel("Iteration")	2✔
67	ax2.set_ylabel("Step Size Std")	2✔
68	ax2.set_title("Step Size Variability")	2✔
69	ax2.grid(True, alpha=0.3)	2✔
70
71	plt.tight_layout()	2✔
72	plt.savefig(f"{outdir}/step_size_evolution.png")	2✔
73
74	except Exception as e:	×
75	print(f"Error generating adaptive MCMC diagnostics: {e}")	×
76
77
78	# NUTS-specific plots
79	## ... unsure what diagnostics we should add...

nz-gravity / LogPSplinePSD / 17934599025

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous