Python List Index Error and Shapes error AND R task fail WHEN REPRODUCING

[dan1elee]

Hi @GjjvdBurg
Thank you for the benchmark work. It's a great work and very thorough. But I'm having some problems reproducing this work. Can I ask if it would be convenient for you to take a very short time to give me some advice on what might be wrong? I'd appreciate it if you could help. The problems I have encountered are as follows.

Python

I ran into a lot of Python errors, the representative traceback are:

Traceback (most recent call last):
File "/home/---/miniconda3/envs/cpd/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
self.run()
File "/home/---/miniconda3/envs/cpd/lib/python3.9/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/home/---/TCPDBench/execs/python/cpdbench_rbocpdms.py", line 64, in wrapper
detector = run_rbocpdms(*args, **kwargs)
File "/home/---/TCPDBench/execs/python/cpdbench_rbocpdms.py", line 144, in run_rbocpdms
detector.run()
File "/home/---/TCPDBench/execs/python/rbocpdms/detector.py", line 320, in run
self.next_run(self.data[t,:], t+1)
File "/home/---/TCPDBench/execs/python/rbocpdms/detector.py", line 360, in next_run
if self.CPs[t-2][-1][0] != self.CPs[t-3][-1][0]:
IndexError: list index out of range

AND

Traceback (most recent call last):
File "/home/---/miniconda3/envs/cpd/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
self.run()
File "/home/---/miniconda3/envs/cpd/lib/python3.9/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/home/---/TCPDBench/execs/python/cpdbench_rbocpdms.py", line 64, in wrapper
detector = run_rbocpdms(*args, **kwargs)
File "/home/---/TCPDBench/execs/python/cpdbench_rbocpdms.py", line 144, in run_rbocpdms
detector.run()
File "/home/---/TCPDBench/execs/python/rbocpdms/detector.py", line 320, in run
self.next_run(self.data[t,:], t+1)
File "/home/---/TCPDBench/execs/python/rbocpdms/detector.py", line 373, in next_run
self.update_all_joint_log_probabilities(y, t)
File "/home/---/TCPDBench/execs/python/rbocpdms/detector.py", line 1102, in update_all_joint_log_probabilities
model.update_joint_log_probabilities(
File "/home/---/TCPDBench/execs/python/rbocpdms/probability_model.py", line 260, in update_joint_log_probabilities
self.update_alpha_derivatives(y, t,
File "/home/---/TCPDBench/execs/python/rbocpdms/probability_model.py", line 634, in update_alpha_derivatives
full = (self.one_step_ahead_predictive_log_loss +
ValueError: operands could not be broadcast together with shapes (5,) (0,)

I'm also experiencing some WARNINGs, not sure if it's related to this.

/home/---/TCPDBench/execs/python/bocpdms/probability_model.py:526: RuntimeWarning: overflow encountered in exp

/home/---/TCPDBench/execs/python/rbocpdms/BVAR_NIG_DPD.py:1553: RuntimeWarning: invalid value encountered in scalar add

AND

/home/---/TCPDBench/execs/python/rbocpdms/BVAR_NIG.py:2876: RuntimeWarning: divide by zero encountered in log

Personally, I think it could be that for some reason some variables are not being assigned or updated successfully, but I verified that the dataset is complete.

R

I also found that some R tasks failed. The json files of the result are:

 {
        "error" : "error with pred probs",
        "command" : "/home/---/libs/R/lib/R/bin/exec/R --no-echo --no-restore --no-save --slave --file=/home/---/TCPDBench/execs/R/cpdbench_ocp.R --args -i /home/---/TCPDBench/datasets/apple.json -l 50 --prior-a 100 --prior-b 0.01 --prior-k 0.01",
        "script" : "/home/---/TCPDBench/execs/R/cpdbench_ocp.R",
        "script_md5" : "909dde17ff91cf60fd649fd6e75ac2fd",
        "hostname" : "ironhide",
        "dataset" : "apple",
        "dataset_md5" : "22edb48471bd3711f7a6e15de6413643",
        "status" : "FAIL",
        "parameters" : {
                "missPts" : "none",
                "cpthreshold" : 0.5,
                "truncRlim" : 0.0001,
                "minRlength" : 1,
                "maxRlength" : 10000,
                "minsep" : 1,
                "maxsep" : 10000,
                "multivariate" : true,
                "lambda" : 50,
                "prior_a" : 100,
                "prior_b" : 0.01,
                "prior_k" : 0.01
        },
        "result" : {
                "cplocations" : null,
                "runtime" : null
        }
}