Integer embeddings in PyTorch
In this video, we implement a paper called "Learning Mathematical Properties of Integers". Most notably, we use an LSTM network and an Encyclopedia of integer sequences to train custom integer embeddings. At the same time, we also extract integer sequences from already pretrained models - BERT and GloVe. We then compare how good these embeddings are at encoding mathematical properties of integers (like divisibility by 2 and primality).
Paper: https://arxiv.org/abs/2109.07230
Code from this video: https://github.com/jankrepl/mildlyoverfitted/tree/master/github_adventures/integer
00:00 Intro
00:41 Ideas and high level explanation
02:56 Data - On-line encyclopedia of Integer Sequences
03:58 Data - raw download exploration
05:43 CustomDataset - implementation
09:15 CustomDataset - testing it out
11:36 Network - implementation
15:54 Network - testing it out
16:58 Evaluation utilities
19:05 GloVe embeddings parsing
22:09 BERT embeddings parsing
24:32 LSTM training script
30:58 Experiments to be run
31:39 Results: LSTM guess next
33:53 Results: Metrics (TensorBoard)
36:33 Results: Embeddings projections (TensorBoard)
40:05 Outro
Видео Integer embeddings in PyTorch канала mildlyoverfitted
Paper: https://arxiv.org/abs/2109.07230
Code from this video: https://github.com/jankrepl/mildlyoverfitted/tree/master/github_adventures/integer
00:00 Intro
00:41 Ideas and high level explanation
02:56 Data - On-line encyclopedia of Integer Sequences
03:58 Data - raw download exploration
05:43 CustomDataset - implementation
09:15 CustomDataset - testing it out
11:36 Network - implementation
15:54 Network - testing it out
16:58 Evaluation utilities
19:05 GloVe embeddings parsing
22:09 BERT embeddings parsing
24:32 LSTM training script
30:58 Experiments to be run
31:39 Results: LSTM guess next
33:53 Results: Metrics (TensorBoard)
36:33 Results: Embeddings projections (TensorBoard)
40:05 Outro
Видео Integer embeddings in PyTorch канала mildlyoverfitted
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![BentoML SageMaker deployment](https://i.ytimg.com/vi/Zci_D4az9FU/default.jpg)
![Retrieval augmented generation with OpenSearch and reranking](https://i.ytimg.com/vi/OsE7YcDcPz0/default.jpg)
![Named entity recognition (NER) model evaluation](https://i.ytimg.com/vi/70YAUYP3hrw/default.jpg)
![Asynchronous requests and rate limiting (HTTPX and asyncio.Semaphore)](https://i.ytimg.com/vi/luWsr9exlE4/default.jpg)
![Few-shot text classification with prompts](https://i.ytimg.com/vi/AhqgDXcBU2M/default.jpg)
![OpenAI function calling](https://i.ytimg.com/vi/_B7F_6nTVEg/default.jpg)
![Deploying machine learning models on Kubernetes](https://i.ytimg.com/vi/DQRNt8Diyw4/default.jpg)
![Haiku basics (neural network library from DeepMind)](https://i.ytimg.com/vi/yXCKS-ZoYTY/default.jpg)
![Product quantization in Faiss and from scratch](https://i.ytimg.com/vi/PNVJvZEkuXo/default.jpg)
![GPT in PyTorch](https://i.ytimg.com/vi/d7IRM40VMYM/default.jpg)
![The Lottery Ticket Hypothesis and pruning in PyTorch](https://i.ytimg.com/vi/bQt0CLXXAqg/default.jpg)
![The Sensory Neuron as a Transformer in PyTorch](https://i.ytimg.com/vi/mi_mzlhBGAU/default.jpg)
![PonderNet in PyTorch](https://i.ytimg.com/vi/JLFz1dU5HR4/default.jpg)
![Mixup in PyTorch](https://i.ytimg.com/vi/hGAKHKqmXdY/default.jpg)
![DINO in PyTorch](https://i.ytimg.com/vi/psmMEWKk4Uk/default.jpg)
![MLP-Mixer in Flax and PyTorch](https://i.ytimg.com/vi/HqytB2GUbHA/default.jpg)
![Differentiable augmentation for GANs (using Kornia)](https://i.ytimg.com/vi/J97EM3Clyys/default.jpg)
![Growing neural cellular automata in PyTorch](https://i.ytimg.com/vi/21ACbWoF2Oo/default.jpg)
![SIREN in PyTorch](https://i.ytimg.com/vi/s4iFEoNlYhM/default.jpg)
![Vision Transformer in PyTorch](https://i.ytimg.com/vi/ovB0ddFtzzA/default.jpg)