Ch3nYe
Ch3nYe
Base on, I already have a GCN model that could generate node data embeddings from graph.
> @resuly @yueshengbin Hello, has your problem been solved? I have the same problem： /opt/conda/lib/python3.7/site-packages/thop/vision/basic_hooks.py:92: UserWarning: **floordiv** is deprecated, and its behavior will change in a future version of pytorch....
I have the same problem, I frozen decoder, only train the encoder. model of 220m could get 0.8+ performance, while model of 2b only get 0.2+ performance. Is the "shallow...
> Adding the following flag worked for me: > > ``` > vllm serve ... --enable-chunked-prefill=False > ``` work for me, on v100s, vllm docker: vllm-openai-v0.6.3