pyrocko.gf.store_ext.StoreExtError: INDEX_OUT_OF_BOUNDS
sxyzy1016 opened this issue · comments
Hi there,
I was trying to calculate the geometry Rec-source using geodetic InSAR data as a practice. I was just following the Example 3 Laquila in the document but without waveform data and just geodetic data.
What I did was:
beat init some_proj- modified the yaml config, and
beat import some_proj, andgeodetic_data.pklwas created succefully. beat build_gfs some_proj --datatypes=geodeticandbeat build_gfs some_proj --force --execute, and everything seems ok.- modified the yaml config, then
beat some_proj --parameters="hierarchicals,hypers"
And then in the follow step beat sample turkey2023 --hypers I run into pyrocko.gf.store_ext.StoreExtError: INDEX_OUT_OF_BOUNDS. The full trace is here:
config - INFO All hierarchicals ok!
config - INFO All hyperparameters ok!
config - INFO All priors ok!
models - INFO ... Initialising Geometry Optimizer ...
models - INFO Analysing problem ...
models - INFO ---------------------
geodetic - INFO Number of geodetic datasets: 1
geodetic - INFO Number of geodetic data points: 1273
geodetic - INFO Initialising corrections ...
heart - INFO Setting up Ramps correction for insar6_2
heart - INFO Not correcting insar6_2 for Euler Pole
heart - INFO Not correcting insar6_2 for Strain Rate
models - INFO ... Building Hyper model ...
geodetic - INFO Evaluating config for Ramps corrections for datasets...
geodetic - INFO Evaluating config for Euler Pole corrections for datasets...
geodetic - INFO No Euler Pole correction!
geodetic - INFO Evaluating config for Strain Rate corrections for datasets...
geodetic - INFO No Strain Rate correction!
geodetic - INFO Initialized 3 hierarchical parameters.
models - INFO Initialised hyperparameter h_SAR with size 1
models - INFO Optimization for 1 hyperparameters in total!
geodetic - INFO Retrieving geodetic data-covariances with structure "import" for insar6_2 ...
Traceback (most recent call last):
File "/home/sxyzy/.conda/envs/beat/bin/beat", line 33, in <module>
sys.exit(load_entry_point('beat==1.2.5', 'console_scripts', 'beat')())
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/apps/beat.py", line 2408, in main
globals()["command_" + command](args)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/apps/beat.py", line 1027, in command_sample
problem = load_model(project_dir, options.mode, options.hypers)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/models/problems.py", line 947, in load_model
problem.built_hyper_model()
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/models/problems.py", line 295, in built_hyper_model
composite.update_llks(point)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/models/geodetic.py", line 462, in update_llks
results = self.assemble_results(point)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/models/geodetic.py", line 224, in assemble_results
processed_synts = self.get_synthetics(point, outmode="stacked_arrays")
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/models/geodetic.py", line 740, in get_synthetics
engine=self.engine, targets=self.targets, sources=self.sources, **kwargs
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/beat-1.2.5-py3.7-linux-x86_64.egg/beat/heart.py", line 4270, in geo_synthetics
response = engine.process(sources, targets)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/pyrocko/gf/seismosizer.py", line 5470, in process
nthreads=nthreads):
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/pyrocko/gf/seismosizer.py", line 4972, in process_static
source, target, components, nthreads)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/pyrocko/gf/seismosizer.py", line 5315, in base_statics
nthreads)
File "/home/sxyzy/.conda/envs/beat/lib/python3.7/site-packages/pyrocko/gf/store.py", line 2084, in statics
nthreads)
pyrocko.gf.store_ext.StoreExtError: INDEX_OUT_OF_BOUNDS
I thought it was a problem of my data format, so I compared my geodetic_data.pkl with the example 3 in spyder, but I got no idea.
I am newbie to this study area. Really need your help... OTL
The final config yaml used in this step was:
--- !beat.BEATconfig
name: turkey2023
date: dummy
event: !pf.Event
time: '1970-01-01 00:00:00'
depth: 10000.0
project_dir: /mnt/d/Proj/Coulomb/turkey2023/turkey2023
problem_config: !beat.ProblemConfig
mode: geometry
source_type: RectangularSource
stf_type: HalfSinusoid
decimation_factors:
geodetic: 4
n_sources: 1
datatypes:
- geodetic
hyperparameters:
h_SAR: !beat.heart.Parameter
name: h_SAR
form: Uniform
lower:
- -2.0
upper:
- 6.0
testvalue:
- 2.0
priors:
depth: !beat.heart.Parameter
name: depth
form: Uniform
lower:
- 0.0
upper:
- 10.0
testvalue:
- 1.0
dip: !beat.heart.Parameter
name: dip
form: Uniform
lower:
- 45.0
upper:
- 90.0
testvalue:
- 63.0
east_shift: !beat.heart.Parameter
name: east_shift
form: Uniform
lower:
- -10.0
upper:
- 10.0
testvalue:
- -8.0
length: !beat.heart.Parameter
name: length
form: Uniform
lower:
- 5.0
upper:
- 30.0
testvalue:
- 11.0
north_shift: !beat.heart.Parameter
name: north_shift
form: Uniform
lower:
- -10.0
upper:
- 10.0
testvalue:
- -8.0
opening_fraction: !beat.heart.Parameter
name: opening_fraction
form: Uniform
lower:
- 0.0
upper:
- 0.0
testvalue:
- 0.0
rake: !beat.heart.Parameter
name: rake
form: Uniform
lower:
- -90.0
upper:
- 90.0
testvalue:
- -72.0
slip: !beat.heart.Parameter
name: slip
form: Uniform
lower:
- 0.1
upper:
- 8.0
testvalue:
- 1.7000000000000002
strike: !beat.heart.Parameter
name: strike
form: Uniform
lower:
- 0.0
upper:
- 180.0
testvalue:
- 36.0
width: !beat.heart.Parameter
name: width
form: Uniform
lower:
- 5.0
upper:
- 20.0
testvalue:
- 9.0
hierarchicals:
insar6_2_azimuth_ramp: !beat.heart.Parameter
name: insar6_2_azimuth_ramp
form: Uniform
lower:
- -0.005
upper:
- 0.005
testvalue:
- 0.0
insar6_2_offset: !beat.heart.Parameter
name: insar6_2_offset
form: Uniform
lower:
- -0.05
upper:
- 0.05
testvalue:
- 0.0
insar6_2_range_ramp: !beat.heart.Parameter
name: insar6_2_range_ramp
form: Uniform
lower:
- -0.005
upper:
- 0.005
testvalue:
- 0.0
geodetic_config: !beat.GeodeticConfig
types:
SAR: !beat.SARDatasetConfig
datadir: /mnt/d/Proj/Coulomb/turkey2023/insar6/merged
names:
- insar6_2
noise_estimator: !beat.GeodeticNoiseAnalyserConfig
structure: import
max_dist_perc: 0.2
interpolation: multilinear
corrections_config: !beat.GeodeticCorrectionsConfig
euler_poles:
- !beat.EulerPoleConfig
enabled: false
ramp: !beat.RampConfig
dataset_names:
- insar6_2
enabled: true
strain_rates:
- !beat.StrainRateConfig
enabled: false
dataset_specific_residual_noise_estimation: false
gf_config: !beat.GeodeticGFConfig
store_superdir: /mnt/d/Proj/Coulomb/turkey2023/turkey2023/BEAT
reference_model_idx: 0
n_variations:
- 0
- 1
earth_model_name: ak135-f-continental.f
nworkers: 20
use_crust2: false
replace_water: false
source_depth_min: 0.0
source_depth_max: 10.0
source_depth_spacing: 1.0
source_distance_radius: 100.0
source_distance_spacing: 1.0
error_depth: 0.1
error_velocities: 0.1
depth_limit_variation: 600.0
version: ''
code: psgrn
sample_rate: 1.1574074074074073e-05
sampling_interval: 1.0
medium_depth_spacing: 1.0
medium_distance_spacing: 10.0
sampler_config: !beat.SamplerConfig
name: SMC
backend: csv
progressbar: true
buffer_size: 5000
buffer_thinning: 1
parameters: !beat.SMCConfig
tune_interval: 50
check_bnd: true
rm_flag: false
n_jobs: 10
n_steps: 100
n_chains: 1000
coef_variation: 1.0
stage: 0
proposal_dist: MultivariateNormal
update_covariances: false
hyper_sampler_config: !beat.SamplerConfig
name: Metropolis
backend: csv
progressbar: true
buffer_size: 5000
buffer_thinning: 1
parameters: !beat.MetropolisConfig
tune_interval: 50
proposal_dist: Normal
check_bnd: true
rm_flag: false
n_jobs: 10
n_steps: 25000
n_chains: 20
thin: 2
burn: 0.5
Hi! Please see BEAT FAQ here: https://pyrocko.org/beat/docs/current/faq.html
Your issues is number 5/6, which are the same but for different data types. From what I can see in your case - it is at least the deoth grid, so please increase source_depth_max to at least 30km, better give a little buffer maybe 35km. Depending on your data and the distance of your furthest data point to the source, you may also need to increase source_distance_radius. Up and better higher than the longest distance.
Cheers! Hannes
Hi!
Thanks a lot for your advice. And sorry for my late response.
I tried lots of times these days, increasing source_depth_max and source_distance_radius, even an extremly high value like 300 and 1000, but the issue still occured. The INDEX_OUT_OF_BOUNDS poped up immediately after the log geodetic - INFO Retrieving geodetic data-covariances with structure "import" for insar6_2.
I also tried decreasing the upper value of depth in the prior, but no help.
And my InSAR deformation data is like:
Please see the last two sentences of my previous response.
Please re-open in case issue still exists.