a/ebooks
1
1
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
91 commits 1 branch 0 tags 164 KiB
Commit graph

91 commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Anthony Wang
dc56b0c327
Fix bug where bot post isn't created with --yes flag 2022-07-19 10:35:47 -05:00
Anthony Wang
40bf86aaff
Make bot.py Matrix backend quit after 5 seconds 2022-07-18 13:44:49 -05:00
Anthony Wang
36baee4925
Rewrite bot.py to rerun generations and allow editing prompt 2022-07-17 23:10:13 -05:00
Anthony Wang
37093e5282
Add more fun prompts 2022-07-17 16:46:17 -05:00
Anthony Wang
458143cf45
Actually remove mentions 2022-07-17 15:50:29 -05:00
Anthony Wang
a4765e17c5
Don't post when instance is empty 2022-07-17 15:16:38 -05:00
Anthony Wang
b96409d739
Remove unneeded import 2022-07-17 10:49:19 -05:00
Anthony Wang
7720c1c161
Update bot.py to use gpt-large 2022-07-17 10:48:11 -05:00
Anthony Wang
e61a793dd6
Adjust training parameters to train gpt2-large 2022-07-16 20:20:34 -05:00
Anthony Wang
d47afd47a3
Adjust training parameters 2022-07-16 11:17:56 -05:00
Anthony Wang
47407b9fb6
Use gpt2-large instead of distilgpt2 2022-07-15 18:48:19 -05:00
Anthony Wang
354ebba789
Use single-quotes instead of double quotes 2022-07-15 13:44:44 -05:00
Anthony Wang
fc56d2c0a8
Make sure output has multiple lines if post isn't long enough 2022-03-23 15:50:56 -05:00
Anthony Wang
3de96f721a
Add a no-op none backend for bot.py 2022-03-01 08:46:01 -06:00
Anthony Wang
81a13d37fb
Add background info to the README 2022-03-01 08:23:09 -06:00
Anthony Wang
3bc9e434ba
Add Matrix support 🎉 2022-02-28 22:23:05 -06:00
Anthony Wang
6513d3369e
Switch to AGPL license 2022-02-28 20:43:44 -06:00
Anthony Wang
7bbd56155b
WIP: Add Pleroma and Misskey support 2022-02-27 18:00:22 -06:00
Anthony Wang
1a13ba9909
Revert "Fix pip command to install pytorch instead of old torch package" 
This reverts commit 123846ed7d.
 2022-02-26 12:06:21 -06:00
Anthony Wang
123846ed7d
Fix pip command to install pytorch instead of old torch package 2022-02-26 12:02:44 -06:00
Anthony Wang
4a9671c958
Remove @ mentions 2022-02-23 15:31:02 -06:00
Anthony Wang
cc60315e21
Make sure that the randomly chosen line has at least two words 2022-02-23 15:08:42 -06:00
Anthony Wang
b93fb7b777
Add more fun initial inputs 2022-02-23 10:15:16 -06:00
Anthony Wang
e7abb12d65
More detailed usage instructions 2022-02-23 10:09:31 -06:00
Anthony Wang
94bd858609
Add flag for bot.py data 2022-02-23 10:07:50 -06:00
Anthony Wang
4c0afade32
Make the bot produce longer posts 2022-02-23 09:59:33 -06:00
Anthony Wang
7f5929b7d0
Make initial input phrases more generic 2022-02-23 07:30:30 -06:00
Anthony Wang
3010b24685
Increase maximum word count 2022-02-22 22:44:17 -06:00
Anthony Wang
758998d17e
Randomly sample data for initial input 50% of the time 2022-02-22 22:42:00 -06:00
Anthony Wang
cd0426ab9e
Pick a random initial input if none provided 2022-02-22 22:25:03 -06:00
Anthony Wang
e1e88500eb
Update README with better bot usage instructions 2022-02-22 22:01:37 -06:00
Anthony Wang
5e8f3591ca
Make sure post is at least 100 characters 2022-02-22 22:01:25 -06:00
Anthony Wang
7f6be5c2ac
Add instance flag for bot 2022-02-22 18:00:10 -06:00
Anthony Wang
f08e5bfc5f
Reformat code with autopep8 2022-02-22 17:55:04 -06:00
Anthony Wang
8dab77d61b
Set output directory correctly 2022-02-22 17:51:52 -06:00
Anthony Wang
1c43115cd6
Split data into chunks and save model 2022-02-22 17:43:29 -06:00
Anthony Wang
cf2d4f1d9b
Add run_clm.py link to README 2022-02-22 17:42:39 -06:00
Anthony Wang
6bab795fe8
Add new transformers bot script and move old one to bot_lstm.py 2022-02-22 16:58:19 -06:00
Anthony Wang
edd4708123
Modify data generation script for transformers 2022-02-22 16:57:12 -06:00
Anthony Wang
a4dc9c238b
Add some transformers documentation to README 2022-02-22 16:17:38 -06:00
Anthony Wang
d191b6204f
Move train.py to train_lstm.py and add new transformers training code 2022-02-22 16:12:13 -06:00
Anthony Wang
fb9b81284b
Optimize LSTM training 2022-02-22 12:36:52 -06:00
Anthony Wang
772f3c3692
Add a resources section to README with useful LSTM links 2022-02-21 22:45:36 -06:00
Anthony Wang
289b8b4bcb
Clean up statuses better in data.py 2022-02-21 19:43:02 -06:00
Anthony Wang
1c1a518fc7
Make each line a status instead of just a big text file of words 2022-02-21 19:20:38 -06:00
Anthony Wang
10c21fb1cd
Only use English messages and clean up URLs better in data.py 2022-02-21 17:57:32 -06:00
Anthony Wang
5f9292e242
Save entire model after training 2022-02-21 16:47:59 -06:00
Anthony Wang
3de6db5dc7
Implement model loading for bot.py 2022-02-21 16:40:27 -06:00
Anthony Wang
6c7935489c
Fix args formatting for train.py 2022-02-21 16:40:09 -06:00
Anthony Wang
b240317d76
Take top 5 instead of top 3 2022-02-21 16:39:58 -06:00