Not enough memory while running fla-t5-base model

Week 1 : Lab 1 - Generative AI Use Case: Summarize Dialogue I am facing issue while running command model_name=‘google/flan-t5-base’ model = AutoModelForSeq2SeqLM.from_pretrained(model_name) with not enough memory error.

I keep getting the following timeout error at the same point, i.e., loading the flan-t5-base model. The lab assignment due time is tonight. Can someone please help?

TimeoutError Traceback (most recent call last) File /opt/conda/lib/python3.11/site-packages/urllib3/connectionpool.py:404, in HTTPConnectionPool._make_request(self, conn, method, url, timeout, chunked, **httplib_request_kw) 403 try: → 404 self._validate_conn(conn) 405 except (SocketTimeout, BaseSSLError) as e: 406 # Py2 raises this as a BaseSSLError, Py3 raises it as socket timeout.

Hi, and welcome to the Forum! Kindly watch this topic for updates regarding this bug. Thanks!

Hey Chris,

I am getting the below error now as I was running into issue with lab1 and I trying to complete it even though the due date has passed but it is not allowing me with the below error.

Hi,

Try in incognito browser mode. In the next time please use a separate thread for your issue.

1 Like

Thank you for your response. I was not able to reply in a separate thread as there were no ‘Reply’ option, the thread shared by Chris was closed probably that’s why it was not allowing any reply. Please share a proper working thread.

Thanks!

Hi Hasnain! Please tell us how you landed on that error. What are the exact steps you took so we can try to reproduce the issue. If possible, please take screenshots at each major step. Thanks.

Thank you Chris for reaching out.

These are the steps I am following.

  1. Launching app by going to Lab 1.

  1. Click Start lab.

  2. Once the lab starts and AWS yellow dot turns Green, I click on AWS.

  1. Looks like it is working in incognito mode because earlier I was not able to get to this page and it shows invalid credentials and contact administrator.

Thank you Chris and Igor, appreciate it.

Thanks!
Hasnain

2 Likes

Great! Glad it works now! Will note the bug you encountered and see if it happens to others too. Will escalate if needed. Thanks.