You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
However, when I reached Step 3: Evaluate the performance of generated text (refer to helm/command/eval.sh), I encountered the following error message:
Traceback (most recent call last):
File "/root/miniconda3/envs/myKV4/lib/python3.8/site-packages/retrying.py", line 251, in call
attempt = Attempt(fn(*args, **kwargs), attempt_number, False)
File "/root/miniconda3/envs/myKV4/lib/python3.8/site-packages/helm/proxy/clients/auto_client.py", line 170, in make_request_with_retry
return client.make_request(request)
File "/root/miniconda3/envs/myKV4/lib/python3.8/site-packages/helm/proxy/clients/together_client.py", line 138, in make_request
raise TogetherClientError("togetherApiKey not set in credentials.conf")
helm.proxy.clients.together_client.TogetherClientError: togetherApiKey not set in credentials.conf
Request failed. Retrying (attempt #2) in 0 seconds... (See above for error details)
Traceback (most recent call last):
File "/root/miniconda3/envs/myKV4/lib/python3.8/site-packages/retrying.py", line 251, in call
attempt = Attempt(fn(*args, **kwargs), attempt_number, False)
File "/root/miniconda3/envs/myKV4/lib/python3.8/site-packages/helm/proxy/clients/auto_client.py", line 170, in make_request_with_retry
return client.make_request(request)
File "/root/miniconda3/envs/myKV4/lib/python3.8/site-packages/helm/proxy/clients/together_client.py", line 138, in make_request
raise TogetherClientError("togetherApiKey not set in credentials.conf")
helm.proxy.clients.together_client.TogetherClientError: togetherApiKey not set in credentials.conf
Request failed. Retrying (attempt #3) in 10 seconds... (See above for error details)
......
I've read the docs, it seems an API key is required. I’m confused, though—aren’t we supposed to be doing an offline evaluation? Why do we need to connect to certain websites?
please advise If there is any way to fix it, thx!
The text was updated successfully, but these errors were encountered:
I followed the instructions in the section Evaluation on tasks from HELM framework(https://github.com/FMInference/H2O/blob/main/h2o_hf/README.md).
However, when I reached
Step 3: Evaluate the performance of generated text (refer to helm/command/eval.sh)
, I encountered the following error message:I've read the docs, it seems an API key is required. I’m confused, though—aren’t we supposed to be doing an offline evaluation? Why do we need to connect to certain websites?
please advise If there is any way to fix it, thx!
The text was updated successfully, but these errors were encountered: