New course announcement ✨

We're teaching an in-person LLM bootcamp in the SF Bay Area on November 14, 2023. Come join us if you want to see the most up-to-date materials building LLM-powered products and learn in a hands-on environment.

[ Ссылка ]

Hope to see some of you there!

--------------------------------------------------------------------------------------------- In this lab, you will upgrade your text recognizer using a ConvNet + LSTM architecture with CTC loss.

Corresponding README: [ Ссылка ].

00:00 - Introduction.
01:59 - Introduce LineCNNSimple, a model that can read multiple characters in an image.
15:52 - Make this model more efficient with LineCNN, which uses a fully convolutional network.
18:18 - Upgrade the model further into LitModelCTC, which uses a CTC (Connectionist Temporal Classification) loss.
23:29 - Finalize your model, LineCNNLSTM, by adding an LSTM layer on top of CNN.
27:34 - Lab 3 homework.

свернуть