KRkristoo·Mar 24, 2021Submission
Show HN: Backprop – a simple library to use and finetune state-of-the-art models
5 points · Open discussionbackprop-ai
backprop
Python2021–2021
backprop-ai / backprop
backprop
Backprop makes it simple to use, finetune, and deploy state-of-the-art ML models.
2Distinct people
4Mentions
0Explicit recommenders
240GitHub stars
What to know before exploring
What it is
Backprop makes it simple to use, finetune, and deploy state-of-the-art ML models.
GitHub
Where it is used
No documented product use has been verified yet.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 3:03:16 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by hour4 dated mentions
Observed mentions: 2021-03-23 15:00:30 UTC to 2021-03-24 14:28:25 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| Mar 23, 15:00 | 2021-03-23 15:00:00 UTC | 2021-03-23 16:00:00 UTC | 1 | 1 | 0 |
| Mar 23, 16:00 | 2021-03-23 16:00:00 UTC | 2021-03-23 17:00:00 UTC | 2 | 0 | 2 |
| Mar 23, 17:00 | 2021-03-23 17:00:00 UTC | 2021-03-23 18:00:00 UTC | 0 | 0 | 0 |
| Mar 23, 18:00 | 2021-03-23 18:00:00 UTC | 2021-03-23 19:00:00 UTC | 0 | 0 | 0 |
| Mar 23, 19:00 | 2021-03-23 19:00:00 UTC | 2021-03-23 20:00:00 UTC | 0 | 0 | 0 |
| Mar 23, 20:00 | 2021-03-23 20:00:00 UTC | 2021-03-23 21:00:00 UTC | 0 | 0 | 0 |
| Mar 23, 21:00 | 2021-03-23 21:00:00 UTC | 2021-03-23 22:00:00 UTC | 0 | 0 | 0 |
| Mar 23, 22:00 | 2021-03-23 22:00:00 UTC | 2021-03-23 23:00:00 UTC | 0 | 0 | 0 |
| Mar 23, 23:00 | 2021-03-23 23:00:00 UTC | 2021-03-24 00:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 00:00 | 2021-03-24 00:00:00 UTC | 2021-03-24 01:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 01:00 | 2021-03-24 01:00:00 UTC | 2021-03-24 02:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 02:00 | 2021-03-24 02:00:00 UTC | 2021-03-24 03:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 03:00 | 2021-03-24 03:00:00 UTC | 2021-03-24 04:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 04:00 | 2021-03-24 04:00:00 UTC | 2021-03-24 05:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 05:00 | 2021-03-24 05:00:00 UTC | 2021-03-24 06:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 06:00 | 2021-03-24 06:00:00 UTC | 2021-03-24 07:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 07:00 | 2021-03-24 07:00:00 UTC | 2021-03-24 08:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 08:00 | 2021-03-24 08:00:00 UTC | 2021-03-24 09:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 09:00 | 2021-03-24 09:00:00 UTC | 2021-03-24 10:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 10:00 | 2021-03-24 10:00:00 UTC | 2021-03-24 11:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 11:00 | 2021-03-24 11:00:00 UTC | 2021-03-24 12:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 12:00 | 2021-03-24 12:00:00 UTC | 2021-03-24 13:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 13:00 | 2021-03-24 13:00:00 UTC | 2021-03-24 14:00:00 UTC | 0 | 0 | 0 |
| Mar 24, 14:00 | 2021-03-24 14:00:00 UTC | 2021-03-24 15:00:00 UTC | 1 | 1 | 0 |
What Hacker News says
4 sourcesYou might get some really promising results with finetuning.
If anything, you could build writing assistance that almost automates responses.
I've been co-authoring a library that lets you finetune such models in a single line of code.
https://github.com/backprop-ai/backprop
In specific the text generation finetuning example should be what you are looking for: https://github.com/backprop-ai/backprop/blob/main/examples/F...
Hope this helps, happy to chat more about it. Pretty curious about the results.
In GPT Neo: open-source GPT model, with pretrained 1.3B & 2.7B weight modelsThat would be one incredibly expensive Q&A system. As rightly suggested by others, some search + a finetuned Q&A system is what you need.
I've been building a library that makes solving tasks like conversational Q&A easy.
https://github.com/backprop-ai/backprop
If you need some pointers, we have example notebooks and I'm happy to chat about it as well.
Would love to know if it helps!
In GPT Neo: open-source GPT model, with pretrained 1.3B & 2.7B weight modelsShow HN: Backprop – a library to easily finetune and use state-of-the-art models
4 points · Open discussion