 Aleksandr Borzunov
|
dc813c426e
Add `src/petals/cli`
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
43e13e1a12
Don't import the client into top-level petals package
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
3ae859858f
Fix running speed test
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
24a60985c7
Fix project_name
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
86efab1fed
Remove duplicate `src`
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
e8b2091a96
Use `python -m petals.cli.run_server` instead of `python -m
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
3ce3fc296d
Move `cli` => `src/petals/cli`
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
67f96d49cf
Fix scripts and notebooks
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
aab5c78e94
Update classifiers
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
eb628f6dfd
Try installing hivemind==1.1.3
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
9c871a987a
Try listing deps in setup.cfg
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
754b6e4745
Update pyproject.toml and setup.cfg
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
bb71bdfa65
Move `petals` => `src/petals`
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
ac22efb8b6
Specify petals-dev@googlegroups.com as an author
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
59db85174e
Fix imports
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
625d042d0a
Add package metadata to pyproject.toml
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
b6ebffe3f8
Expose BloomTokenizerFast in petals.client
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
d1c35b4f5f
Rename `src` => `petals`
|
%!s(int64=2) %!d(string=hai) anos |
 Alexander Borzunov
|
0c3781a89c
Shorten bullet points in readme
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
ab41223b17
Fix dtype- and device-related client issues (#98)
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
c6e1b5a8e5
Add various server timeouts, lower --max_batch_size and --inference_max_length defaults (#97)
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
d8ef09146e
Improve server's logging (#96)
|
%!s(int64=2) %!d(string=hai) anos |
 Artem Chumachenko
|
fdb3583a8c
Add Beam Search decoding algorithm (#87)
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
fef7257fe0
Try to fix protobuf versions once again (#95)
|
%!s(int64=2) %!d(string=hai) anos |
|
Aleksandr Borzunov
|
1b51703444
Revert protobuf version change
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
b26b0b7121
Require hivemind with fixed compression and protobuf working on Colab (#94)
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
8a73b41a42
Make ServerState announcements work better (#93)
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
dc71574a63
Use public swarm by default (#92)
|
%!s(int64=2) %!d(string=hai) anos |
|
Alexander Borzunov
|
11d6ba683c
Make inference, forward, and backward fully fault-tolerant (#91)
|
%!s(int64=2) %!d(string=hai) anos |
|
Artem Chumachenko
|
695df826c2
Force reinstall for hivemind in example notebooks (#88)
|
%!s(int64=2) %!d(string=hai) anos |
