Step map_your_entity_network failed

Operating System

• macOS
• Linux
• Windows

Deployment Method

• Docker
• non-Docker

CUDA Usage

• Yes
• No

Training Process Details (if applicable)

1. Model Size: qwen3: 0.6b
2. Machine base configuration: RAM 8GB, Disk 256GB

Second-Me version

latest master

Describe the bug

Map entity network failed: Command '['bash', '/Users/shuheng-mo/Git Workspaces/Second-Me/lpm_kernel/L2/data_pipeline/data_prep/scripts/graphrag_indexing.sh']' returned non-zero exit status 1. 2025-05-25 00:59:03 [INFO] trainprocess_service.py:390 - Cleaning up resources to prevent memory leaks 2025-05-25 00:59:04 [INFO] trainprocess_service.py:404 - Memory usage after cleanup: 313.58 MB 2025-05-25 00:59:04 [ERROR] trainprocess_service.py:1112 - Step map_your_entity_network failed 2025-05-25 00:59:04 [INFO] trainprocess_service.py:1113 - Marking step as failed: stage=map_your_entity_network, step=map_your_entity_network 2025-05-25 01:03:19 [INFO] routes.py:204 - Progress reset successfully

Current Behavior

I tried reset training but still failed at 'map_your_entity_network' step.

Expected Behavior

Expected it to run successfully, since the RAM is not likely to leak according to the doc, if so please give me some guidance.

Reproduction Steps

By hitting "Start Training" in Train Second Me will reproduce it.

Possible Workaround

I see there is a previous issue about this, but it was not closed so I think I'd better reminds

Additional Information

None

Link to related Github discussion or issue

Issue 348