Deep Conclusions

51 posts

Deep Conclusions banner
Deep Conclusions

Deep Conclusions

@deepconclusions

We develop Ai products and teach people to code

Kampala, Uganda Joined Kasım 2022
71 Following16 Followers
Deep Conclusions retweeted
Mike Driscoll
Mike Driscoll@driscollis·
#Python Pop Quiz 🐍❓ What is the output of this silly code? A) NameError B) 0 C) TypeError D) None of the above
Mike Driscoll tweet media
English
10
4
40
6.2K
Deep Conclusions retweeted
LlamaIndex 🦙
LlamaIndex 🦙@llama_index·
When building GraphRAG, you may want to explicitly define the graph yourself, or use the LLM automatically extract the graph. Both have tradeoffs: the former takes a lot more human effort but gives you full control over the data representation, and the latter automates human effort but can be unreliable. With @llama_index, you can do either or both! 🌟 Our extractor can do the following: 1. Parse documents into nodes 2. Extract paths from text using an LLM - either free-form or in accordance to a schema. 3. Define relationships manually using our Node abstractions (you can link different nodes) - it’ll auto convert to graph relationships 4. Generate embeddings for every node Check out our full guide here: docs.llamaindex.ai/en/stable/exam…
LlamaIndex 🦙 tweet media
English
5
89
372
34K
Deep Conclusions retweeted
Yann LeCun
Yann LeCun@ylecun·
The No Language Left Behind paper (NLLB) just appeared in Nature. High-quality translation between 200 languages in any direction, with sparse training data, and many low-resource languages.
AI at Meta@AIatMeta

Newly published today in @Nature: No Language Left Behind (NLLB) is an AI model created by researchers at Meta capable of delivering high-quality translations directly between 200 languages – including low-resource languages. Read more in Nature ⬇️ go.fb.me/7j8o4p

English
118
260
2.5K
357.7K
Deep Conclusions retweeted
Google DeepMind
Google DeepMind@GoogleDeepMind·
🔍 Analysts need to rewatch many game replays to study rival teams and design future tactics. TacticAI can help by automatically computing numerical representations of players, allowing them to efficiently look up relevant past routines. ↓ dpmd.ai/49PGq1b
English
1
6
91
14.9K
Deep Conclusions retweeted
Mike Driscoll
Mike Driscoll@driscollis·
#Python Pop Quiz 🐍❓ What is the output of the following code? A) True B) False C) (7, 8, 9) D) None of the above
Mike Driscoll tweet media
English
16
4
38
5.7K
Deep Conclusions retweeted
PyTorch
PyTorch@PyTorch·
Check out what's new in PyTorch Documentation 👀 Featuring 3 new tutorials and PyTorch 2.2 documentation + tutorials! Details 👉 hubs.la/Q02jwnLt0
PyTorch tweet media
English
4
26
136
22.7K
Deep Conclusions retweeted
Mike Driscoll
Mike Driscoll@driscollis·
Did you know that #Python supports SQLite in the standard library? That's right. You don't need to install anything! 🐍🚀 Let's talk a little about that! 🧵👇
English
9
19
134
22.6K