How to run analytical lpa_u2i on the test dataset (lpa_3000_xx)?

[andreich78-20]

The question
Hi,
I am trying to run the code very similar to what we have in unittests and getting the error:
CompilationError: Failed to compile 27e6afdfeb36fa629ee77fa943ca6f6aa42c019af9188f4e31c8caf7e784f385 on platform
system: Linux

Steps

1. Cloned the test datasets repo with git clone -b master --single-branch --depth=1 https://github.com/7br/gstest.git GraphScopeTestData
2. Copied the files from "/property/lpa_dataset" to "/Lab"
3. Invoked the code (I did my best to be close to unittest)

import pandas as pd
import graphscope

import os
cwd = os.getcwd()

graphscope.set_option(show_log=True)
sess = graphscope.session(cluster_type="hosts", num_workers=2, mode="eager")

g = sess.g(generate_eid=False, retain_oid=False)
g = g.add_vertices(f"{cwd}/Lab/lpa_3000_v_0", "v0")
g = g.add_vertices(f"{cwd}/Lab/lpa_3000_v_1", "v1")
g = g.add_edges(
    f"{cwd}/Lab/lpa_3000_e_0", "e0", ["weight"], "v0", "v1"
)

property_context = graphscope.lpa_u2i(g, max_round=20)
#print(property_context.schema)
df = property_context.to_dataframe(
    {"node": "v:v0.id", "label0": "r:v0.label_0", "label1": "r:v0.label_1"}
)
print("dataframe:")
print(df)


sess.close()

Expected behavior
Code should go to the end, lpa_u2i should run as it does during the tests.

Environment (please complete the following information):

• GraphScope version: 0.28.0
• OS: WSL2 on Windows 10
• Kubernetes Version v1.25.2

Attachments
log.txt

PS
I have digged all around in search for GraphScope custom lpa example, only unit tests are available - but no luck extracting the code...