深度碎片

icon indicating a website link https://forums.fast.ai/t/fast-ai-v3-2019/39325

icon location wuhan, Hubei, China icon location 4th week on fastHTML

Results 116 issues of 深度碎片

Different ways to put your data into DataBunch

### homework for week 1 91:00-91:35 ### Different ways to put your data into DataBunch 91:35-95:10 [Different ways to put your data into DataBunch](https://ytcropper.com/cropped/Xf5c6c9c4c91af3) How to use MNIST sample dataset?...

How to improve model with more layers?

### How to improve model with more layers? 86:55-91:00 [How to improve model with more layers](https://ytcropper.com/cropped/Xf5c6c957044c1a) to use ResNet50 instead of ResNet34 - `data = ImageDataBunch.from_name_re(path_img, fnames, pat, ds_tfms=get_transforms(), size=299,...

How to train the whole model in the right way?

### How to train the whole model in the right way? 82:32- 86:55 [How to train the whole model in the right way](https://ytcropper.com/cropped/Xf5c6c90ca38d85) How we go back to the unbroken...

what is CNN actually learning and why previous full model training didn’t work

### what is CNN actually learning and why previous full model training didn’t work 76:24 - 82:32 [what is CNN actually learning and why previous full model training didn’t work](https://ytcropper.com/cropped/Xf5c6c894589a28)...

How to improve our model? finetuning

### How to improve our model? finetuning 74:37 - 76:26 [How to improve our model](https://ytcropper.com/cropped/Xf5c6c848ee9bd2) What is the default way of training? - add a few layers at the end...

How to find out the most confused images of our model?

### How to find out the most confused images of our model? 73:20 -74:39 [How to find out the most confused images of our model](https://ytcropper.com/cropped/Xf5c6c80043d55e) * why we need the...

how to plot top losses examples/images?

### how to plot top losses examples/images? 68:43-73:22 [how to plot top losses examples/images](https://ytcropper.com/cropped/Xf5c6c77fc6fc4e) How to create model interpreter? - `interp = ClassificationInterpretation.from_learner(learn)` Why plot the high loss? - to...

How to save a trained model?

### How to save a trained model? 67:57-68:43 [How to save a trained model](https://ytcropper.com/cropped/Xf5c6be210b8cbd) What is inside the trained model? - updated weights why do we need to save model?...

Why use Resnet rather than Inception?

### Why use Resnet rather than Inception? 66:00 - 67:57 [Why use Resnet rather than Inception](https://ytcropper.com/cropped/Xf5c6be0e24a093) * DAWNBench on ImageNet classification * “Resnet is good enough” for top 5 places...

What students achieved with fastai and this course?

### What students achieved with fastai and this course? 52:45-65:51 [What students achieved with fastai and this course](https://ytcropper.com/cropped/Xf5c6bdc2e86b00) Sarah Hooker - first course student, economics (no background in coding) -...