Commit Graph

155 Commits

Author SHA1 Message Date
oobabooga
f3da6dcc8f
Merge pull request #149 from oobabooga/RWKV
Add RWKV support
2023-03-01 16:57:45 -03:00
oobabooga
831ac7ed3f Add top_p 2023-03-01 16:45:48 -03:00
oobabooga
7c4d5ca8cc Improve the text generation call a bit 2023-03-01 16:40:25 -03:00
oobabooga
2f16ce309a Rename a variable 2023-03-01 12:33:09 -03:00
oobabooga
9e9cfc4b31 Parameters 2023-03-01 12:19:37 -03:00
oobabooga
0f6708c471 Sort the imports 2023-03-01 12:18:17 -03:00
oobabooga
e735806c51 Add a generate() function for RWKV 2023-03-01 12:16:11 -03:00
oobabooga
659bb76722 Add RWKVModel class 2023-03-01 12:08:55 -03:00
oobabooga
9c86a1cd4a Add RWKV pip package 2023-03-01 11:42:49 -03:00
oobabooga
6837d4d72a Load the model by name 2023-02-28 02:52:29 -03:00
oobabooga
a1429d1607 Add default extensions to the settings 2023-02-28 02:20:11 -03:00
oobabooga
19ccb2aaf5 Handle <USER> and <BOT> 2023-02-28 01:05:43 -03:00
oobabooga
626da6c731 Handle {{user}} and {{char}} in example dialogue 2023-02-28 00:59:05 -03:00
oobabooga
e861e68e38 Move the chat example dialogue to the prompt 2023-02-28 00:50:46 -03:00
oobabooga
f871971de1 Trying to get the chat to work 2023-02-28 00:25:30 -03:00
oobabooga
67ee7bead7 Add cpu, bf16 options 2023-02-28 00:09:11 -03:00
oobabooga
ebd698905c Add streaming to RWKV 2023-02-28 00:04:04 -03:00
oobabooga
70e522732c Move RWKV loader into a separate file 2023-02-27 23:50:16 -03:00
oobabooga
ebc64a408c RWKV support prototype 2023-02-27 23:03:35 -03:00
oobabooga
021bd55886 Better format the prompt when generation attempts > 1 2023-02-27 21:37:03 -03:00
oobabooga
43b6ab8673 Store thumbnails as files instead of base64 strings
This improves the UI responsiveness for large histories.
2023-02-27 13:41:00 -03:00
oobabooga
f24b6e78a3 Fix clear history 2023-02-26 23:58:04 -03:00
oobabooga
8e3e8a070f Make FlexGen work with the newest API 2023-02-26 16:53:41 -03:00
oobabooga
3333f94c30 Make the gallery extension work on colab 2023-02-26 12:37:26 -03:00
oobabooga
633a2b6be2 Don't regenerate/remove last message if the chat is empty 2023-02-26 00:43:12 -03:00
oobabooga
6e843a11d6 Fix FlexGen in chat mode 2023-02-26 00:36:04 -03:00
oobabooga
4548227fb5 Downgrade gradio version (file uploads are broken in 3.19.1) 2023-02-25 22:59:02 -03:00
oobabooga
9456c1d6ed Prevent streaming with no_stream + generation attempts > 1 2023-02-25 17:45:03 -03:00
oobabooga
32f40f3b42 Bump gradio version to 3.19.1 2023-02-25 17:20:03 -03:00
oobabooga
fa58fd5559 Proper way to free the cuda cache 2023-02-25 15:50:29 -03:00
oobabooga
b585e382c0 Rename the custom prompt generator function 2023-02-25 15:13:14 -03:00
oobabooga
700311ce40 Empty the cuda cache at model.generate() 2023-02-25 14:39:13 -03:00
oobabooga
1878acd9f3 Minor bug fix in chat 2023-02-25 09:30:59 -03:00
oobabooga
e71ff959f5 Clean up some unused code 2023-02-25 09:23:02 -03:00
oobabooga
91f5852245 Move bot_picture.py inside the extension 2023-02-25 03:00:19 -03:00
oobabooga
5ac24b019e Minor fix in the extensions implementation 2023-02-25 02:53:18 -03:00
oobabooga
85f914b9b9 Disable the hijack after using it 2023-02-25 02:36:01 -03:00
oobabooga
7e9f13e29f Rename a variable 2023-02-25 01:55:32 -03:00
oobabooga
1741c36092 Minor fix 2023-02-25 01:47:25 -03:00
oobabooga
7c2babfe39 Rename greed to "generation attempts" 2023-02-25 01:42:19 -03:00
oobabooga
2dfb999bf1 Add greed parameter 2023-02-25 01:31:01 -03:00
oobabooga
13f2688134 Better way to generate custom prompts 2023-02-25 01:08:17 -03:00
oobabooga
67623a52b7 Allow for permanent hijacking 2023-02-25 00:55:19 -03:00
oobabooga
111b5d42e7 Add prompt hijack option for extensions 2023-02-25 00:49:18 -03:00
oobabooga
7a527a5581 Move "send picture" into an extension
I am not proud of how I did it for now.
2023-02-25 00:23:51 -03:00
oobabooga
e51ece21c0 Add ui() function to extensions 2023-02-24 19:00:11 -03:00
oobabooga
78ad55641b Remove duplicate max_new_tokens parameter 2023-02-24 17:19:42 -03:00
oobabooga
65326b545a Move all gradio elements to shared (so that extensions can use them) 2023-02-24 16:46:50 -03:00
oobabooga
0817fe1beb Move code back into the chatbot wrapper 2023-02-24 14:10:32 -03:00
oobabooga
8a7563ae84 Reorder the imports 2023-02-24 12:42:43 -03:00
oobabooga
ace74a557a Add some comments 2023-02-24 12:41:27 -03:00
oobabooga
fe5057f932 Simplify the extensions implementation 2023-02-24 10:01:21 -03:00
oobabooga
2fb6ae6970 Move chat preprocessing into a separate function 2023-02-24 09:40:48 -03:00
oobabooga
f6f792363b Separate command-line params by spaces instead of commas 2023-02-24 08:55:09 -03:00
oobabooga
e260e84e5a Merge branch 'max_memory' of https://github.com/elwolf6/text-generation-webui into elwolf6-max_memory 2023-02-24 08:47:01 -03:00
oobabooga
146f786c57 Reorganize a bit 2023-02-24 08:44:54 -03:00
oobabooga
c2f4c395b9 Clean up some chat functions 2023-02-24 08:31:30 -03:00
luis
5abdc99a7c gpu-memory arg change 2023-02-23 18:43:55 -05:00
oobabooga
9ae063e42b Fix softprompts when deepspeed is active (#112) 2023-02-23 20:22:47 -03:00
oobabooga
dac6fe0ff4 Reset the history if no default history exists on reload 2023-02-23 19:53:50 -03:00
oobabooga
3b8cecbab7 Reload the default chat on page refresh 2023-02-23 19:50:23 -03:00
oobabooga
f1914115d3 Fix minor issue with chat logs 2023-02-23 16:04:47 -03:00
oobabooga
b78561fba6 Minor bug fix 2023-02-23 15:26:41 -03:00
oobabooga
2e86a1ec04 Move chat history into shared module 2023-02-23 15:11:18 -03:00
oobabooga
c87800341c Move function to extensions module 2023-02-23 14:55:21 -03:00
oobabooga
2048b403a5 Reorder functions 2023-02-23 14:49:02 -03:00
oobabooga
7224343a70 Improve the imports 2023-02-23 14:41:42 -03:00
oobabooga
e46c43afa6 Move some stuff from server.py to modules 2023-02-23 13:42:23 -03:00
oobabooga
1dacd34165 Further refactor 2023-02-23 13:28:30 -03:00
oobabooga
ce7feb3641 Further refactor 2023-02-23 13:03:52 -03:00
oobabooga
98af4bfb0d Refactor the code to make it more modular 2023-02-23 12:05:25 -03:00
oobabooga
bc856eb962 Add some more margin 2023-02-20 20:49:21 -03:00
oobabooga
f867285e3d Make the circle a bit less red 2023-02-20 18:41:38 -03:00
oobabooga
e4440cd984 Make highlighted text gray in cai-chat mode 2023-02-20 16:43:32 -03:00
oobabooga
995bcfcf5e Minor style change 2023-02-18 22:14:57 -03:00
oobabooga
d58544a420 Some minor formatting changes 2023-02-18 11:07:55 -03:00
oobabooga
3e6a8ccdce Fix galactica latex css 2023-02-18 00:18:39 -03:00
oobabooga
14f49bbe9a Fix galactica equations in dark mode 2023-02-17 23:57:09 -03:00
oobabooga
abb4667b44 Improve basic HTML style 2023-02-17 23:08:34 -03:00
oobabooga
00ca17abc9 Minor change 2023-02-17 22:52:03 -03:00
oobabooga
2fd003c044 Fix gpt4chan styles that were broken by gradio 3.18.0 2023-02-17 22:47:41 -03:00
oobabooga
0dd41e4830 Reorganize the sliders some more 2023-02-17 16:33:27 -03:00
oobabooga
6b9ac2f88e Reorganize the generation parameters 2023-02-17 16:18:01 -03:00
oobabooga
3923ac967f Create a cache for profile pictures (in RAM)
This is a performance optimization.
2023-02-17 14:30:39 -03:00
oobabooga
a6ddbbfc77 Add more fonts options 2023-02-17 11:30:04 -03:00
oobabooga
5eeb3f4e54 Make thumbnails for the profile pictures (for performance) 2023-02-17 10:58:54 -03:00
oobabooga
71c2764516 Fix the API docs in chat mode 2023-02-17 01:56:51 -03:00
oobabooga
33ad21c4f2 Make the profile pictures a bit larger 2023-02-17 00:35:17 -03:00
oobabooga
c4e87c109e Include the bot's image as base64
This is needed for Colab.
2023-02-17 00:24:27 -03:00
oobabooga
aeddf902ec Make the refresh button prettier 2023-02-16 21:55:20 -03:00
oobabooga
3746d72853 More style fixes 2023-02-15 21:13:12 -03:00
oobabooga
a55e8836f6 Bump gradio version
It looks uglier, but the old one was bugged and unstable.
2023-02-15 20:20:56 -03:00
oobabooga
1622059179 Move BLIP to the CPU
It's just as fast
2023-02-15 00:03:19 -03:00
oobabooga
8c3ef58e00 Use BLIP directly + some simplifications 2023-02-14 23:55:46 -03:00
SillyLossy
a7d98f494a Use BLIP to send a picture to model 2023-02-15 01:38:21 +02:00
oobabooga
56bbc996a4 Minor CSS change for readability 2023-02-13 23:01:14 -03:00
oobabooga
61aed97439 Slightly increase a margin 2023-02-12 17:38:54 -03:00
oobabooga
76d3d7ddb3 Reorder the imports here too 2023-02-10 15:57:55 -03:00
oobabooga
d038963193 Rename a variable (for #59) 2023-02-07 23:26:02 -03:00
oobabooga
f38c9bf428 Fix deepspeed (oops) 2023-02-02 10:39:37 -03:00