transformers: Upload models using Git fails
Environment info
transformersversion: 3.5.0- Platform: Linux-4.15.0-112-generic-x86_64-with-debian-buster-sid
- Python version: 3.6.12
- PyTorch version (GPU?): 1.7.0 (False)
- Tensorflow version (GPU?): 2.3.1 (False)
- Using GPU in script?: No
- Using distributed or parallel set-up in script?: No
Who can help
Model Cards: @julien-c T5: @patrickvonplaten
Information
Model I am using (Bert, XLNet …): T5
The problem arises when using:
- the official example scripts: (give details below)
- my own modified scripts: (give details below)
The tasks I am working on is:
- an official GLUE/SQUaD task: (give the name)
- my own task or dataset: (give details below)
To reproduce
Steps to reproduce the behavior:
git clone https://huggingface.co/Rostlab/prot_t5_xl_bfd
Cloning into 'prot_t5_xl_bfd'...
remote: Enumerating objects: 31, done.
remote: Counting objects: 100% (31/31), done.
remote: Compressing objects: 100% (29/29), done.
remote: Total 31 (delta 13), reused 0 (delta 0), pack-reused 0
Unpacking objects: 100% (31/31), done.
cp config.json pytorch_model.bin prot_t5_xl_bf
git add --all
git status
On branch main
Your branch is up to date with 'origin/main'.
Changes to be committed:
(use "git reset HEAD <file>..." to unstage)
modified: config.json
modified: pytorch_model.bin
git commit -m "[T5] Fix load weights function #8528"
git push
Username for 'https://huggingface.co': xxxxxx
Password for 'https://agemagician@huggingface.co':
Counting objects: 4, done.
Delta compression using up to 80 threads.
Compressing objects: 100% (3/3), done.
Writing objects: 100% (4/4), 4.92 GiB | 23.09 MiB/s, done.
Total 4 (delta 2), reused 1 (delta 0)
error: RPC failed; HTTP 504 curl 22 The requested URL returned error: 504 Gateway Time-out
fatal: The remote end hung up unexpectedly
fatal: The remote end hung up unexpectedly
Everything up-to-date
OR
GIT_CURL_VERBOSE=1 git push
* Couldn't find host huggingface.co in the .netrc file; using defaults
* Trying 192.99.39.165...
* TCP_NODELAY set
* Connected to huggingface.co (192.99.39.165) port 443 (#0)
* found 140 certificates in /etc/ssl/certs/ca-certificates.crt
* found 421 certificates in /etc/ssl/certs
* ALPN, offering http/1.1
* SSL connection using TLS1.2 / ECDHE_RSA_AES_256_GCM_SHA384
* server certificate verification OK
* server certificate status verification SKIPPED
* common name: huggingface.co (matched)
* server certificate expiration date OK
* server certificate activation date OK
* certificate public key: RSA
* certificate version: #3
* subject: CN=huggingface.co
* start date: Tue, 10 Nov 2020 08:05:46 GMT
* expire date: Mon, 08 Feb 2021 08:05:46 GMT
* issuer: C=US,O=Let's Encrypt,CN=Let's Encrypt Authority X3
* compression: NULL
* ALPN, server accepted to use http/1.1
> GET /Rostlab/prot_t5_xl_bfd/info/refs?service=git-receive-pack HTTP/1.1
Host: huggingface.co
User-Agent: git/2.17.1
Accept: */*
Accept-Encoding: gzip
Accept-Language: C, *;q=0.9
Pragma: no-cache
< HTTP/1.1 401 Unauthorized
< Server: nginx/1.14.2
< Date: Sat, 14 Nov 2020 23:43:52 GMT
< Content-Type: text/plain; charset=utf-8
< Content-Length: 12
< Connection: keep-alive
< X-Powered-By: huggingface-moon
< WWW-Authenticate: Basic realm="Authentication required", charset="UTF-8"
< ETag: W/"c-dAuDFQrdjS3hezqxDTNgW7AOlYk"
<
* Connection #0 to host huggingface.co left intact
Username for 'https://huggingface.co': agemagician
Password for 'https://agemagician@huggingface.co':
* Couldn't find host huggingface.co in the .netrc file; using defaults
* Found bundle for host huggingface.co: 0x55acdab63f80 [can pipeline]
* Re-using existing connection! (#0) with host huggingface.co
* Connected to huggingface.co (192.99.39.165) port 443 (#0)
* Server auth using Basic with user 'agemagician'
> GET /Rostlab/prot_t5_xl_bfd/info/refs?service=git-receive-pack HTTP/1.1
Host: huggingface.co
Authorization: Basic YWdlbWFnaWNpYW46VWRjXzEyMDA=
User-Agent: git/2.17.1
Accept: */*
Accept-Encoding: gzip
Accept-Language: C, *;q=0.9
Pragma: no-cache
< HTTP/1.1 200 OK
< Server: nginx/1.14.2
< Date: Sat, 14 Nov 2020 23:43:59 GMT
< Content-Type: application/x-git-receive-pack-advertisement
< Transfer-Encoding: chunked
< Connection: keep-alive
< X-Powered-By: huggingface-moon
<
* Connection #0 to host huggingface.co left intact
Counting objects: 4, done.
Delta compression using up to 80 threads.
Compressing objects: 100% (3/3), done.
* Couldn't find host huggingface.co in the .netrc file; using defaults
* Found bundle for host huggingface.co: 0x55acdab63f80 [can pipeline]
* Re-using existing connection! (#0) with host huggingface.co
* Connected to huggingface.co (192.99.39.165) port 443 (#0)
* Server auth using Basic with user 'agemagician'
> POST /Rostlab/prot_t5_xl_bfd/git-receive-pack HTTP/1.1
Host: huggingface.co
Authorization: Basic YWdlbWFnaWNpYW46VWRjXzEyMDA=
User-Agent: git/2.17.1
Content-Type: application/x-git-receive-pack-request
Accept: application/x-git-receive-pack-result
Content-Length: 4
* upload completely sent off: 4 out of 4 bytes
< HTTP/1.1 200 OK
< Server: nginx/1.14.2
< Date: Sat, 14 Nov 2020 23:44:02 GMT
< Content-Type: application/x-git-receive-pack-result
< Content-Length: 0
< Connection: keep-alive
< X-Powered-By: huggingface-moon
<
* Connection #0 to host huggingface.co left intact
* Couldn't find host huggingface.co in the .netrc file; using defaults
* Found bundle for host huggingface.co: 0x55acdab63f80 [can pipeline]
* Re-using existing connection! (#0) with host huggingface.co
* Connected to huggingface.co (192.99.39.165) port 443 (#0)
* Server auth using Basic with user 'xxxxxx'
> POST /Rostlab/prot_t5_xl_bfd/git-receive-pack HTTP/1.1
Host: huggingface.co
Authorization: Basic YWdlbWFnaWNpYW46VWRjXzEyMDA=
User-Agent: git/2.17.1
Accept-Encoding: gzip
Content-Type: application/x-git-receive-pack-request
Accept: application/x-git-receive-pack-result
Transfer-Encoding: chunked
Writing objects: 100% (4/4), 4.92 GiB | 23.17 MiB/s, done.
Total 4 (delta 2), reused 1 (delta 0)
* Signaling end of chunked upload via terminating chunk.
* Signaling end of chunked upload via terminating chunk.
* The requested URL returned error: 504 Gateway Time-out
* stopped the pause stream!
* Closing connection 0
error: RPC failed; HTTP 504 curl 22 The requested URL returned error: 504 Gateway Time-out
fatal: The remote end hung up unexpectedly
fatal: The remote end hung up unexpectedly
Everything up-to-date
Expected behavior
I had issue before for uploading big models like T5 “#7480”. It was marked as solved after moving to model versioning: https://github.com/huggingface/transformers/pull/8324 However, @patrickvonplaten fixed some problems in T5 weights : https://github.com/huggingface/transformers/pull/8528 I tried to update the model but it still doesn’t work as showed above.
I tried several tricks like: https://stackoverflow.com/questions/54061758/error-rpc-failed-http-504-curl-22-the-requested-url-returned-error-504-gatewa But could not solve it.
Any ideas ?
About this issue
- Original URL
- State: closed
- Created 4 years ago
- Comments: 20 (8 by maintainers)
Running into the same issue here trying to upload a 1.4Gb dataset
Having the same issue when uploading a ~6GB GPT-J model via command line:
It seems to be always around 5.11GB (total file size is 6.32GB). When uploading the same file via the HF website interface I get an error 400. Tried it from various locations (wifis), always the same behavior.
Any helpful advises on how to proceed?
For anyone how still couldn’t figured it out: it is because the big file is not being tracked git LFS. Reasons maybe:
.gitattributes.gitattributes, but the file extension is not in the listYou can simply ask git LFS to track your file, for example:
git lfs track "*.gguf"@sgugger @thomwolf can you reopen this issue?
This happened to me too:
Me too
@julien-c @patrickvonplaten uploading a mbart model from google colab and got it . any pointers ??
No worries at all! Feel free to open a new issue -> doesn’t take me long at all to upload it 😃