Dog Breed Identification challenge

(Sujay Madbhavi) #162

Thanks @KevinB will report it.

(Sujay Madbhavi) #163

yeah works fine now after that fix Thanks @rob

(Shikhar Gupta) #164

Can someone help me with this error. I’m not able to find any post related to this. I’m trying to replicate the dogbreed notebook. I’m trying to run this on AWS p2 instance.

(Kevin Bird) #165

Get rid of that line I believe is what Jeremy told somebody else. Other thing you could try is setting it to 0 but I think that’s the default.

(Shikhar Gupta) #166

Thanks I just saw it in the beginner forum

(Jeremy Howard) #167

Oh apologies I didn’t realize that it’s not in 0.2. I’ll revert.

(Phani Srikanth) #168

The tqdm progress bar seems to be broken for me.
Is anyone facing this formatting issue?

(Jeremy Howard) #169

That just means you need to restart your kernel. It’s a bug in tqdm, or probably actually in the jupyter console code.

(Phani Srikanth) #170

Damn! Works like a charm now :sweat_smile:

(Miguel Perez Michaus) #171

That was quick! It works fine now :slight_smile:

(Chris Palmer) #172

Trying to understand if this conversation is about how to train with the entire data and all of the intelligence of the resnext model - what is meant by not “removing the last layer from pretrianed models” and " they are using the fully connected layers where we take those off and calculate our own"?

How can I do what is being talked about here - what is it that we do that removes the last layer, what can we do that preserves it?

Also, how can we retain all of the data to train with?

(yinterian) #173

If you just want to remove the last layer you may need to do a little bit of extra work. One possibility is what I did here.

This could be simplified with the new API that Jeremy just wrote. See an example here.

I am happy to help if you have any questions. My example work for vgg16 you would have to understand the network that you are trying to change.

(Jeremy Howard) #174

I’ve got an example of how to cut layers off custom models that I’ll be pushing soon-ish. We’ll be discussing it later in the course.

(Jeremy Howard) #175

OK this is at the more advanced end for now - but for those interested in understanding fastai more deeply:

I just added nasnet.ipynb . It shows how to use a new pretrained model type that’s not already in fastai. (This particularly one is really slow BTW, although it should be better with 0.3).

Note that I also changed fastai.models.nasnet to optionally skip that classifier section.

Wiki: Fastai Library Feature Requests
(James Requa) #176

wow nasnet, thats awesome…Thanks @jeremy !

For anyone interested in learning more about it…

(Chris Palmer) #177

Actually I am quite confused about what is being discussed here.

What I was asking here was for someone to explain exactly what were these guys getting at when they said that the (I presumed standard) approach is “removing the last layer from pretrianed models” - because the guys that are NOT doing that are getting better results. I was not asking how to remove the last layer - but how to understand what the contributors @bushaev, @jamesrequa and @KevinB meant that we should NOT be removing it, and that they (the better scoring people) are “using the fully connected layers where we take those off and calculate our own”. Do they mean that we shouldn’t set precompute to false? It didn’t make sense to me…

(Kevin Bird) #178

I was asking for clarification, but if I understand it correctly, they are doing well because the data is literally being trained on the same images that are in the test set. So this would be comparable to training using our validation set. It would give us a very good score, but it doesn’t really help in any real-world scenario. They are basically exploiting the fact that they know what data the set comes from and choosing a model that starts with that and keeping those activations. I was and still am asking for clarification on whether I actually understand why they are doing so well, but that is what I was asking in my post.

(Chris Palmer) #179

I have tried this, and val_idxs = 0, and also val_idxs = get_cv_idxs(n, val_pct=0.01), but I continue to get an error (below). Does anyone know how to do this reliably?

Should we have one image per class in our validation set?

AssertionError                            Traceback (most recent call last)
<ipython-input-24-0708a7145fb8> in <module>()
----> 1 learn = ConvLearner.pretrained(arch, data, precompute=True, ps=0.5)

~/fastai/courses/dl1/fastai/ in pretrained(cls, f, data, ps, xtra_fc, xtra_cut, **kwargs)
     92     def pretrained(cls, f, data, ps=None, xtra_fc=None, xtra_cut=0, **kwargs):
     93         models = ConvnetBuilder(f, data.c, data.is_multi, data.is_reg, ps=ps, xtra_fc=xtra_fc, xtra_cut=xtra_cut)
---> 94         return cls(data, models, **kwargs)
     96     @property

~/fastai/courses/dl1/fastai/ in __init__(self, data, models, precompute, **kwargs)
     85         elif self.metrics is None:
     86             self.metrics = [accuracy_multi] if else [accuracy]
---> 87         if precompute: self.save_fc1()
     88         self.freeze()
     89         self.precompute = precompute

~/fastai/courses/dl1/fastai/ in save_fc1(self)
    132         self.fc_data = ImageClassifierData.from_arrays(,
    133                 (act,, (val_act,,,,
--> 134                 test = test_act if else None, num_workers=8)
    136     def freeze(self):

~/fastai/courses/dl1/fastai/ in from_arrays(cls, path, trn, val, bs, tfms, classes, num_workers, test)
    296             ImageClassifierData
    297         """
--> 298         datasets = cls.get_ds(ArraysIndexDataset, trn, val, tfms, test=test)
    299         return cls(path, datasets, bs, num_workers, classes=classes)

~/fastai/courses/dl1/fastai/ in get_ds(fn, trn, val, tfms, test, **kwargs)
    264     def get_ds(fn, trn, val, tfms, test=None, **kwargs):
    265         res = [
--> 266             fn(trn[0], trn[1], tfms[0], **kwargs), # train
    267             fn(val[0], val[1], tfms[1], **kwargs), # val
    268             fn(trn[0], trn[1], tfms[1], **kwargs), # fix

~/fastai/courses/dl1/fastai/ in __init__(self, x, y, transform)
    160     def __init__(self, x, y, transform):
    161         self.x,self.y=x,y
--> 162         assert(len(x)==len(y))
    163         super().__init__(transform)
    164     def get_x(self, i): return self.x[i]


(Kevin Bird) #180

Can you post your data variable generation too?

(Chris Palmer) #181

PATH = "data/dogbreed/"
tfms = tfms_from_model(arch, sz, aug_tfms=transforms_side_on, max_zoom=1.1)
data = ImageClassifierData.from_csv(PATH, ‘train’, f’{PATH}labels.csv’, bs=bs, tfms=tfms,
val_idxs=val_idxs, suffix = ‘.jpg’, test_name = ‘test’,