Yuge Zhang
Yuge Zhang
The error can be safely ignored. We might address this in future.
That really depends on your definition of agents. For generally information on how to evaluate a trained model with VERL, please refer to https://github.com/volcengine/verl/issues/298
Hi, can you use calc_agent_dev.py to debug?
How did you run it? Detailed traceback please.
How did you run the script? Please provide the command.
I think the agentops is not properly configured even after setting the environment varaibles. Not sure why. Could you get into agentops.client.Client here and print what self.config looks like? https://github.com/AgentOps-AI/agentops/blob/817f554936431dddfc2ae75b30ad252812d63628/agentops/client/client.py#L151
agentops break a lot of times when upgrading. So we pin it to avoid unexpected upgrade. Will upgrade the pin.
Please provide your env info, the command you are running.
Under plan of v0.3