How to run

AMO-ACO Inference, based code here:

AMO-ACO

Change cfg in AMO-ACO.ipynb to config suitable for your Dataset you want to run.

We have 3 options: Data_100, Data_200 and Data_400.

Experimental Result

Average time for 100 node datasets is 400 seconds - 500 epochs.

Name		Name	Last commit message	Last commit date
Latest commit History 38 Commits
Mock_data		Mock_data
Model		Model
__pycache__		__pycache__
data_gen		data_gen
txt		txt
.DS_Store		.DS_Store
AM.txt		AM.txt
AMO-ACO-100-No-Start.pt		AMO-ACO-100-No-Start.pt
AMO-ACO-100.pt		AMO-ACO-100.pt
AMO-ACO-train-20.pt		AMO-ACO-train-20.pt
AMO-ACO.ipynb		AMO-ACO.ipynb
Ant.py		Ant.py
Ant_k_starts.py		Ant_k_starts.py
Check_Solution.txt		Check_Solution.txt
Cross_Exchange.py		Cross_Exchange.py
D-AM.pt		D-AM.pt
D-AM.txt		D-AM.txt
Data.txt		Data.txt
Draw.py		Draw.py
IBSO.py		IBSO.py
Injection.py		Injection.py
Load_data.py		Load_data.py
Local_search.py		Local_search.py
Mock_data.py		Mock_data.py
Normalize_data.py		Normalize_data.py
POMO-MT.txt		POMO-MT.txt
POMO.txt		POMO.txt
README.MD		README.MD
Report.pdf		Report.pdf
So_sanh.txt		So_sanh.txt
Solution.txt		Solution.txt
TW_GPU_Dynamic_Attention_Model.ipynb		TW_GPU_Dynamic_Attention_Model.ipynb
a.ipynb		a.ipynb
diagram.png		diagram.png
illustration.png		illustration.png
mock_data.txt		mock_data.txt
result.png		result.png
ultis.py		ultis.py