Part 2 & Alumni (2018)


Study Groups 2 Many fast.ai students have found that they stick to their study plans and understand the material better if they study together with a group. Such a group can be virtual (hangouts/skype/etc) or in person. It could be folks at your workplace, a meetup of people in your local community, or just other students from the forum that are in your approximate time zone. We’ve seen all of these models work well.
Topic Replies Activity
About the Part 2 & Alumni (2018) category 33 January 30, 2019
Part 2 Lesson 8 wiki 259 May 11, 2020
Train model on GPU and run it on CPU for predictions 4 May 7, 2020
Test Seq2Seq prediction after training 10 April 16, 2020
Project: implement yolo v3 backbone and preact resnet 68 May 5, 2020
Using Pretrained Models from Cadene repository 9 April 26, 2020
ULMFIT - Hungarian 1 April 26, 2020
What does the encoder actually learn? :thinking: 67 April 26, 2020
ULMFiT on Highly imbalanced Datasets 4 April 20, 2020
How can I load a pretrained model on Kaggle using fastai? 36 April 18, 2020
Update Model attribute via Callback 5 April 12, 2020
Tuning hyper-parameters with bayesian optimization 13 April 8, 2020
Lesson 1. Bounding box detection in fastai-1.x 4 April 7, 2020
How to fix 'Error(s) in loading state_dict for AWD_LSTM' when using fast-ai 15 April 7, 2020
Rust or Python 2 April 5, 2020
Build your deep learning box: wiki thread 160 April 2, 2020
Language Model Zoo :gorilla: 407 March 30, 2020
ULMFIT - Persian, Farsi 1 March 30, 2020
ULMfit - during fit, my loss is showing #na# 4 March 28, 2020
Online training of language models 7 March 25, 2020
Reset GPU without restarting linux? 9 March 21, 2020
Study Group in Nigeria 3 March 18, 2020
ULMFIT - Chinese 2 March 5, 2019
Custom Dataloader for ABSA task 4 March 12, 2020
Multiple regression for bbox only in fastai v1 23 March 5, 2020
ULMFiT - Indonesian 12 February 24, 2020
Image classification competition 3 February 22, 2020
Exposing DL models as api's/microservices 35 February 20, 2020
Lottery Ticket Sub-Nets: Is this a real breakthrough? 22 February 19, 2020
Seq2seq: Better to improve spacy or build a custom model for named entity recognition? 7 February 19, 2020