Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

convert-hf : save memory with lazy evaluation #7075

Merged
merged 26 commits into from
May 8, 2024
Merged

convert-hf : save memory with lazy evaluation #7075

Changes from 1 commit
Commits
Show all changes
26 commits
Select commit Hold shift + click to select a range
47e02eb
convert-hf : begin refactoring write_tensor
compilade Apr 30, 2024
0d720ac
Merge branch 'master' into compilade/convert-hf-refactor
compilade Apr 30, 2024
c33775b
convert : upgrade to sentencepiece v0.2.0
compilade Apr 30, 2024
698f0b3
convert-hf : remove unused n_dims in extra_*_tensors
compilade Apr 30, 2024
cde9ea6
convert-hf : simplify MoE weights stacking
compilade Apr 30, 2024
56f60f5
convert-hf : flake8 linter doesn't like semicolons
compilade May 1, 2024
3870164
convert-hf : allow unusual model part names
compilade May 1, 2024
dcd8dfa
convert : use a string for the SentencePiece tokenizer path
compilade May 1, 2024
21068b6
convert-hf : display tensor shape
compilade May 1, 2024
639b374
convert-hf : convert norms to f32 by default
compilade May 1, 2024
644c269
convert-hf : sort model part names
compilade May 1, 2024
ce067af
convert-hf : use an ABC for Model again
compilade May 2, 2024
13f4cf7
convert-hf : use a plain class for Model, and forbid direct instantia…
compilade May 2, 2024
6a54973
Merge branch 'master' into compilade/convert-hf-refactor
compilade May 3, 2024
3e5e0dc
Merge branch 'master' into compilade/convert-hf-refactor
compilade May 3, 2024
98f2d0e
convert-hf : more consistent formatting of cmdline args
compilade May 4, 2024
f2099c5
convert-hf : align the message logged for converted tensors
compilade May 4, 2024
215a0d3
convert-hf : fix Refact conversion
compilade May 5, 2024
f09674f
convert-hf : save memory with lazy evaluation
compilade May 4, 2024
0c38332
convert-hf : flake8 doesn't like lowercase L as a variable name
compilade May 4, 2024
98db434
convert-hf : remove einops requirement for InternLM2
compilade May 4, 2024
bc78bf4
convert-hf : faster model parts loading
compilade May 5, 2024
62303e7
convert-hf : minor changes for consistency
compilade May 5, 2024
68c5ac6
Merge branch 'master' into compilade/lazy-convert-hf
compilade May 6, 2024
94e667a
gguf-py : add tqdm as a dependency
compilade May 6, 2024
bffdaf4
Merge branch 'master' into compilade/lazy-convert-hf
compilade May 8, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Merge branch 'master' into compilade/convert-hf-refactor
  • Loading branch information
@compilade
compilade committed May 3, 2024
commit 6a54973d82f11fd55423081260838345daae433f

This merge commit was added into this branch cleanly.

There are no new changes to show, but you can still view the diff.