oobabooga
b937c9d8c2
Add skip_special_tokens checkbox for Dolly model ( #1218 )
2023-04-16 14:24:49 -03:00
oobabooga
a9c7ef4159
Exclude yaml files from model list
2023-04-16 12:47:30 -03:00
oobabooga
4e035cc3fb
Fix api-example-stream
2023-04-16 12:12:31 -03:00
oobabooga
b705b4210c
Minor changes to training.py
2023-04-16 03:08:37 -03:00
oobabooga
5c513a5f5c
Make training.py more readable
2023-04-16 02:46:27 -03:00
Alex "mcmonkey" Goodwin
a3eec62b50
Lora trainer improvements part 3 ( #1098 )
...
* add support for other model types
dependent on future-peft-changes but with fallback to function now
* use encoding=utf8 for training format
* make shuffling optional
and describe dropout a bit more
* add eval_steps to control evaluation
* make callbacks not depend on globals
* make save steps controllable
* placeholder of initial loading-existing-model support
and var name cleanup
* save/load parameters
* last bit of cleanup
* remove `gptq_bits` ref as main branch removed that setting
* add higher_rank_limit option
2048 is basically unreachable due to VRAM, but i trained at 1536 with batch size = 1 on a 7B model.
Note that it's in the do_train input just to save as a parameter
* fix math on save_steps
2023-04-16 02:35:13 -03:00
kernyan
ac19d5101f
revert incorrect eos_token_id change from #814 ( #1261 )
...
- fixes #1054
2023-04-16 01:47:01 -03:00
oobabooga
a2127239de
Fix a bug
2023-04-16 01:41:37 -03:00
oobabooga
9d3c6d2dc3
Fix a bug
2023-04-16 01:40:47 -03:00
Mikel Bober-Irizar
16a3a5b039
Merge pull request from GHSA-hv5m-3rp9-xcpf
...
* Remove eval of API input
* Remove unnecessary eval/exec for security
* Use ast.literal_eval
* Use ast.literal_eval
---------
Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>
2023-04-16 01:36:50 -03:00
oobabooga
d2ea925fa5
Bump llama-cpp-python to use LlamaCache
2023-04-16 00:53:40 -03:00
oobabooga
ac189011cb
Add "Save current settings for this model" button
2023-04-15 12:54:02 -03:00
oobabooga
b9dcba7762
Don't overwrite --gpu_memory on boot (#1237/#1235)
2023-04-15 11:59:31 -03:00
oobabooga
72a89e8fd1
Merge branch 'main' of github.com:oobabooga/text-generation-webui
2023-04-15 01:21:46 -03:00
oobabooga
abef355ed0
Remove deprecated flag
2023-04-15 01:21:19 -03:00
Lou Bernardi
4d13cc0a1f
Suppress whisper_stt pip warning ( #1215 )
2023-04-15 01:15:59 -03:00
catalpaaa
94700cc7a5
Bump gradio to 3.25 ( #1089 )
2023-04-14 23:45:25 -03:00
oobabooga
004f275efe
Update README.md
2023-04-14 23:36:56 -03:00
oobabooga
83964ed354
Update README.md
2023-04-14 23:33:54 -03:00
oobabooga
c41037db68
Update README.md
2023-04-14 23:32:39 -03:00
oobabooga
f8913c7132
Merge branch 'main' of github.com:oobabooga/text-generation-webui
2023-04-14 23:17:29 -03:00
oobabooga
628f8e6168
Reorganize chat buttons
2023-04-14 23:17:15 -03:00
DavG25
5126c97139
Improve CSS in default and notebook mode ( #1209 )
2023-04-14 23:03:43 -03:00
oobabooga
c3aa79118e
Minor generate_chat_prompt simplification
2023-04-14 23:02:08 -03:00
oobabooga
c4aa1a42b1
Fix chat history downloading
2023-04-14 19:38:30 -03:00
oobabooga
3a337cfded
Use argparse defaults
2023-04-14 15:35:06 -03:00
oobabooga
43e01282b3
Don't override user initial wbits/groupsize
2023-04-14 15:24:03 -03:00
Alex "mcmonkey" Goodwin
64e3b44e0f
initial multi-lora support ( #1103 )
...
---------
Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>
2023-04-14 14:52:06 -03:00
oobabooga
ebb81eb176
Add Undo/Regenerate buttons to notebook mode
2023-04-14 14:34:56 -03:00
oobabooga
1901d238e1
Minor change to API code
2023-04-14 12:11:47 -03:00
oobabooga
581f739b2f
Reorganize
2023-04-14 11:38:26 -03:00
oobabooga
7d7d122edb
Cover one more model
2023-04-14 11:15:59 -03:00
oobabooga
8eba88061a
Remove unused config
2023-04-14 11:12:17 -03:00
oobabooga
8e31f2bad4
Automatically set wbits/groupsize/instruct based on model name ( #1167 )
2023-04-14 11:07:28 -03:00
v0xie
9d66957207
Add --listen-host launch option ( #1122 )
2023-04-13 21:35:08 -03:00
oobabooga
403be8a27f
Update README.md
2023-04-13 21:23:35 -03:00
dependabot[bot]
852a5aa13d
Bump bitsandbytes from 0.37.2 to 0.38.1 ( #1158 )
2023-04-13 21:23:14 -03:00
dependabot[bot]
84576a80d2
Bump llama-cpp-python from 0.1.30 to 0.1.33 ( #1157 )
2023-04-13 21:17:59 -03:00
oobabooga
2908a51587
Settle for transformers 4.28.0
2023-04-13 21:07:00 -03:00
oobabooga
28a11f5724
Revert the Default preset to a previous version
2023-04-13 13:02:02 -03:00
oobabooga
a75e02de4d
Simplify GPTQ_loader.py
2023-04-13 12:13:07 -03:00
oobabooga
c13e8651ad
Suppress "TypedStorage is deprecated" warnings
2023-04-13 12:09:42 -03:00
oobabooga
17ce7c8671
Suppress annoying bitsandbytes welcome message
2023-04-13 12:04:39 -03:00
oobabooga
ca293bb713
Show a warning if two quantized models are found
2023-04-13 12:04:27 -03:00
oobabooga
fbb448ce4f
If only 1 model is available, load that model
2023-04-13 11:44:10 -03:00
oobabooga
8b482b4127
Merge #1073 from sgsdxzy/triton
...
* Multi-GPU support for triton
* Better quantized model filename detection
2023-04-13 11:31:21 -03:00
oobabooga
fde6d06167
Prioritize names with the groupsize in them
2023-04-13 11:27:03 -03:00
oobabooga
f2bf1a2c9e
Add some comments, remove obsolete code
2023-04-13 11:17:32 -03:00
oobabooga
5744b31593
Reorganize some buttons
2023-04-13 11:05:47 -03:00
Light
da74cd7c44
Generalized weight search path.
2023-04-13 21:43:32 +08:00
