bdqnghi / infercode

[ICSE 2021] - InferCode: Self-Supervised Learning of Code Representations by Predicting Subtrees

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

How to initial two InferCodeClient with different languages?

allomy opened this issue · comments

Hi, I tried to use InferCode to generate vectors for java and c languages, and I want to initial two InferCodeClients as the following code shows, but at line 6 infercode_c_predictor.init_from_config() will give the following exception.

from infercode.client.infercode_client import InferCodeClient

infercode_java_predictor = InferCodeClient(language="java")
infercode_java_predictor.init_from_config()
infercode_c_predictor = InferCodeClient(language="c")
infercode_c_predictor.init_from_config()   // Line 6
// Error Message
Exception has occurred: NotFoundError
Restoring from checkpoint failed. This is most likely due to a Variable name or other graph key that is missing from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Key dense_2/kernel not found in checkpoint
	 [[node save_1/RestoreV2 (defined at /home/******/.local/lib/python3.8/site-packages/infercode/client/infercode_client.py:45) ]]

......

So is there anyway that I get two InferCodeClients with different languages at the same time?

Thank you for your help in advance.

Hi, is this repo still being updated?

maybe add "tf.reset_default_graph()" before creating the second InferCodeClient can solve this problem