dhruvt93
(Dhruv Thakur)
November 1, 2018, 10:17am
593
I’m getting the following plot after running learn.lr_find()
. I’ve chosen the lr
range that corresponds to the middle drop. I can see that the valid_loss
is beginning to increase on the 6th epoch, but what can be the reason for a constant error_rate
? Let me know if more context is required.
dhruvt93
(Dhruv Thakur)
November 1, 2018, 10:25am
594
P.S. This is the case when lr
is in range (1e-6,4e-6)
too.
aloksaan
(Alok)
November 1, 2018, 11:19am
595
Updating the fastai library fixed it
marcmuc
(Marc P. Rostock)
November 1, 2018, 11:22am
596
answered here
From what I understand it’s the following.
The model trains on the training set, and the training loss is calculated on that training set. So this loss is calculated on the same dataset the model was trained on and should thus be low.
However, the model has never seen the validation set, so when the validation loss is calculated it is on new data for a model only trained on the training set.
Hence the training loss ought to be lower than the validation set, because the model was fitted on the…
Yes. All non apple and non orange images in training set to be classified as Other.
1 Like
sgugger
November 1, 2018, 1:21pm
599
No it’s not a random direction, but the direction of the gradients. The stochastic refers to the fact we draw batches randomly.
1 Like
Rename then with the most correct labels from a human perspective.
Michael, Have you seen this ?
Again, been following the code to see how a specific problem is solved: i.e. what do you do with images when the input size is not the exact prescribed dimensions (e.g. 224 x 224, 229 x 229 etc)
I think this code is at the kernal of resizing:
def scale_min(im, targ):
""" Scales the image so that the smallest axis is of size targ.
Arguments:
im (array): image
targ (int): target size
"""
r,c = im.size
ratio = targ/min(r,c)
sz = (scale_to(r, ratio, targ), …
1 Like
As Andrew Ng has said, it is like worrying about over population on Mars.
jeremy
(Jeremy Howard)
November 1, 2018, 1:41pm
603
I removed that one since it’s not the approach we recommend.
1 Like
jeremy
(Jeremy Howard)
November 1, 2018, 1:42pm
604
Please read the etiquette guide in the FAQ.
joshfp
(José Fernández Portal)
November 1, 2018, 1:42pm
605
I’m facing the same issue and agree that replacing valid_ds
with train_ds
is not ok. Looking at the code of the ClassificationInterpretation class, it seems that it only works with the validation set. I guess would be great that it would receive a parameter to select the dataset.
1 Like
jeremy
(Jeremy Howard)
November 1, 2018, 1:44pm
607
miwojc:
Question: Image size, size parameter in ImageDataBunch, that is set to 224. Is it better that is you will get lower error when images are higher resolution (it will take more time, bs smaller)? or the image resolution need to be 224 for resnet34?
You will get lower error when using higher res images.
2 Likes
That’s OK. Autoencoders are too noisy anyway, which means the NN learns the quirks of the autoencoder, rather than what really makes an image a member of a class.
In the video, Jeremy has explained this very clearly
vedder
November 1, 2018, 2:01pm
610
Thanks I’ll be sure to check it out. I am using the same dataset so this could be really helpful!
@miwojc has kindly shared this Jeremy paper on another forum. It may help.
1 Like
Sorry. I did not see that you had already shared the same resource. Thanks.