Lesson 1 In-Class Discussion ✅

Part 1 (2019)
533

Jeremy said it defaults to 0.003 if left blank.

1 Like
534

I tested this wrapper on the lesson-1 example before running learner.fit and it works beautifully for running the model on multiple-gpus:

learn = ConvLearner(data, models.resnet34, metrics=error_rate)
learn.model = torch.nn.DataParallel(learn.model)
4 Likes
536

Thank you!

537

look at feature visualisation papers. The first layer will produce a activation map where the activations are high are in their specific locations. The next layer will pass a filter through this, taking neighbour hood activations into account and produce another map of activations and their locations and so on…

538

does FAI does it by default… ,what different does it do

539

Thank you!

540

Thanks for the class and all your work responding to questions!

541

don’t forget to like the video! https://www.youtube.com/watch?v=7hX8yKCX6xM

1 Like
542

Thank you to Jeremy and all involved! :clap::clap::clap:

543

thx a lot

544

Can anyone link to the thread/post of NLT project which Jeremy mentioned?
Thanks

545

Thanks!

546

if we have segmentation problems where we have to read the images and predict the masks… in that case what version of loader can we use…as dont hve any labels for such kind of problem

547

thx all!

548

Great ! thx a lot

549

as jeremy said, multi-cpu by default but only single gpu

550

Thank you, Jeremy!

Big claps!:clap::clap::clap::clap::clap:

1 Like
551

Amazing lecture @jeremy. Looking forward to learn about how to create our own dataset from google images. Thanks!

552

gotta have label, this is Supervised training. Unsupervised training is possible, but that’s in Part 2.

1 Like
553

if we do nn.parallel what different would it do