In Week 1 Lab Assignment,
I am not able to load the huggingface dataset. It is giving readtimeout exception (Attached screenshot)
Tried changing the internal network, but didn’t work.
Any suggestions here?
There may be a temporary issue with the Hugging Face server, please try rerunning the code a bit later.
Exact same issue for me. I hope we can still finish and submit the lab when this is fixed
It seems that a lot of learners are having the same issue right now:
In the thread above someone has suggested a workaround.
In my experience issues with Hugging Face servers are usually resolved quickly, but I am aware of an incident that lasted a couple of days that took place in April of this year.
Of course you will be able to submit the lab after the issue is resolved.
I am encountering the same issue as well:
TimeoutError Traceback (most recent call last)
File /opt/conda/lib/python3.10/site-packages/urllib3/connectionpool.py:467, in HTTPConnectionPool._make_request(self, conn, method, url, body, headers, retries, timeout, chunked, response_conn, preload_content, decode_content, enforce_content_length)
466 try:
→ 467 self._validate_conn(conn)
468 except (SocketTimeout, BaseSSLError) as e:
…
Does anyone know if there is a workaround? Is it possible to download the dataset another way and upload it to the Sagemaker environment manually?
I was able to manually download the CSV files from hugging face and drop them into the project. I created another cell to load the files manually into the dataset dict:
data_files = {‘train’: ‘train.csv’, ‘validation’: ‘validation.csv’, ‘test’: ‘test.csv’}
dataset = load_dataset(‘csv’, data_files=data_files)
Now I am getting an error loading the FLAN-T5 model. I guess we just wait until HF is back up? Will we still be able to do the lab? My understanding is that the lab is only good for 2 hours.
Thank you
Yes you can do the lab again, you will have to open it the same way.
In case you encounter issues with the AWS budget, although I don’t think there will be any, you can fill a Lab issue report (GenAI with LLMs Lab Issue Report).
1 Like
OK thank you. Is there any status page for HF where we can check and see if they are back up? If this is a regular occurrence it might be good to add this to the FAQ. Thank you!
Issue seems to be fixed now !
… correction, it got past downloadoing the readme, but got same error when trying to download the data.
Same issue for me. I also was able to manually load the data files, but getting TimeoutError loading the FLAN_T5 model:
ReadTimeout: (ReadTimeoutError("HTTPSConnectionPool(host='cdn-lfs.hf.co', port=443): Read timed out. (read timeout=10)")
Same issue here. Please help if anyone know the solution.
- Changing Network > tried
Errer when running below cell:
huggingface_dataset_name = “knkarthick/dialogsum”
dataset = load_dataset(huggingface_dataset_name)
Here is the error just the first 10-20 lines
TimeoutError Traceback (most recent call last)
File /opt/conda/lib/python3.10/site-packages/urllib3/connectionpool.py:467, in HTTPConnectionPool._make_request(self, conn, method, url, body, headers, retries, timeout, chunked, response_conn, preload_content, decode_content, enforce_content_length)
466 try:
→ 467 self._validate_conn(conn)
468 except (SocketTimeout, BaseSSLError) as e:
File /opt/conda/lib/python3.10/site-packages/urllib3/connectionpool.py:1099, in HTTPSConnectionPool._validate_conn(self, conn)
1098 if conn.is_closed:
→ 1099 conn.connect()
1101 # TODO revise this, see Emit a warning when proxy_is_verified is False · Issue #2791 · urllib3/urllib3 · GitHub