Comprehensive Image Classification Introduction @fastbook 05
from fastai.vision.all import *
path = untar_data(URLs.PETS)
Path.BASE_PATH = path
path.ls(), (path/"images").ls()
fname = (path/"images").ls()[0]
fname, re.findall(r'(.+)_\d+.jpg$', fname.name)
pets = DataBlock(
blocks = (ImageBlock, CategoryBlock),
get_items=get_image_files,
splitter=RandomSplitter(seed=42),
get_y=using_attr(RegexLabeller(r'(.+)_\d+.jpg$'), 'name'),
item_tfms=Resize(460),
batch_tfms=aug_transforms(size=224, min_scale=0.75)
)
dls = pets.dataloaders(path/"images")
f = using_attr(RegexLabeller(r'(.+)_\d+.jpg$'), 'name')
f(fname)
%cd /notebooks/fastbook
dblock1 = DataBlock(
blocks=(
ImageBlock(),
CategoryBlock()
),
get_y=parent_label,
item_tfms=Resize(460)
)
# Place an image in the 'images/grizzly.jpg' subfolder where this notebook is located before running this
dls1 = dblock1.dataloaders([(Path.cwd()/'images'/'grizzly.jpg')]*100, bs=8)
dls1.train.get_idxs = lambda: Inf.ones
x,y = dls1.valid.one_batch()
_,axs = subplots(1, 2)
x1 = TensorImage(x.clone())
x1 = x1.affine_coord(sz=224)
x1 = x1.rotate(draw=30, p=1.)
x1 = x1.zoom(draw=1.2, p=1.)
x1 = x1.warp(draw_x=-0.2, draw_y=0.2, p=1.)
tfms = setup_aug_tfms([
Rotate(draw=30, p=1, size=224),
Zoom(draw=1.2, p=1., size=224),
Warp(draw_x=-0.2, draw_y=0.2, p=1., size=224)
])
x = Pipeline(tfms)(x)
#x.affine_coord(coord_tfm=coord_tfm, sz=size, mode=mode, pad_mode=pad_mode)
TensorImage(x[0]).show(ctx=axs[0])
TensorImage(x1[0]).show(ctx=axs[1]);
How 'parent_label()' works
fname, parent_label(fname)
Path.cwd(), Path.cwd().ls()
dls.show_batch(nrows=1, ncols=3)
added 'Resize(460)' for GPU, Otherwise 'summary()' doesn't work
pets1 = DataBlock(blocks = (ImageBlock, CategoryBlock),
get_items=get_image_files,
splitter=RandomSplitter(seed=42),
get_y=using_attr(RegexLabeller(r'(.+)_\d+.jpg$'), 'name'),
item_tfms=Resize(460)) # same size for GPU
pets1.summary(path/"images")
learn = vision_learner(dls, resnet34, metrics=error_rate)
learn.fine_tune(2)
dls.vocab[2]
x,y = dls.one_batch()
preds,y1 = learn.get_preds(dl=[(x,y)])
y, preds[0]
len(preds[0]),preds[0].sum()
plot_function(torch.sigmoid, min=-4,max=4)
plot_function(torch.exp, min=-4,max=4)
acts = torch.randn((6,2))*2
acts
acts.sigmoid()
torch.exp(acts[:,0]) / torch.exp(acts).sum(dim=1)
sm_acts = torch.softmax(acts, dim=1)
sm_acts
sm_acts.sum(dim=1)
targ = tensor([0,1,0,1,1,0])
sm_acts
idx = range(6)
sm_acts[idx, targ]
from IPython.display import HTML
df = pd.DataFrame(sm_acts, columns=["3","7"])
df['targ'] = targ
df['idx'] = idx
df['result'] = sm_acts[range(6), targ]
t = df.style.hide_index()
#To have html code compatible with our script
html = t._repr_html_().split('</style>')[1]
html = re.sub(r'<table id="([^"]+)"\s*>', r'<table >', html)
display(HTML(html))
-sm_acts[idx, targ]
F.nll_loss(sm_acts, targ, reduction='none')
Taking the Log
Recall that cross entropy loss may involve the multiplication of many numbers. Multiplying lots of negative numbers together can cause problems like numerical underflow in computers. Therefore, we want to transform these probabilities to larger values so we can perform mathematical operations on them. There is a mathematical function that does exactly this: the logarithm (available as torch.log). It is not defined for numbers less than 0, and looks like this between 0 and 1:
plot_function(torch.log, min=0,max=1, ty='log(x)', tx='x')
plot_function(lambda x: -1*torch.log(x), min=0,max=1, tx='x', ty='- log(x)', title = 'Log Loss when true label = 1')
from IPython.display import HTML
df['loss'] = -torch.log(tensor(df['result']))
t = df.style.hide_index()
#To have html code compatible with our script
html = t._repr_html_().split('</style>')[1]
html = re.sub(r'<table id="([^"]+)"\s*>', r'<table >', html)
display(HTML(html))
loss_func = nn.CrossEntropyLoss()
loss_func(acts, targ)
F.cross_entropy(acts, targ)
nn.CrossEntropyLoss(reduction='none')(acts, targ)
acts, targ
interp = ClassificationInterpretation.from_learner(learn)
interp.plot_confusion_matrix(figsize=(12,12), dpi=60)
interp.most_confused(min_val=5)
learn = vision_learner(dls, resnet34, metrics=error_rate)
learn.fine_tune(1, base_lr=0.1)
learn = vision_learner(dls, resnet34, metrics=error_rate)
lr_min,lr_steep = learn.lr_find(suggest_funcs=(minimum, steep))
print(f"Minimum/10: {lr_min:.2e}, steepest point: {lr_steep:.2e}")
(lr_min + lr_steep)/2.
learn = vision_learner(dls, resnet34, metrics=error_rate)
learn.fine_tune(3, base_lr=3e-3)
learn.recorder.plot_loss()
plt.plot(L((learn.recorder).values).itemgot(2))
learn.fine_tune??
learn = vision_learner(dls, resnet34, metrics=error_rate)
learn.fit_one_cycle(3, 3e-3)
learn.unfreeze()
learn.lr_find()
learn.fit_one_cycle(6, lr_max=1e-5)
learn.recorder.plot_loss()
learn = vision_learner(dls, resnet34, metrics=error_rate)
learn.fit_one_cycle(3, 3e-3)
learn.unfreeze()
learn.fit_one_cycle(12, lr_max=slice(1e-6,1e-4))
plt.plot(L(learn.recorder.values).itemgot(2))
learn.recorder.plot_loss()
from fastai.callback.fp16 import *
learn = vision_learner(dls, resnet50, metrics=error_rate).to_fp16()
learn.fine_tune(6, freeze_epochs=3)
- Why do we first resize to a large size on the CPU, and then to a smaller size on the GPU?
- If you are not familiar with regular expressions, find a regular expression tutorial, and some problem sets, and complete them. Have a look on the book's website for suggestions.
- What are the two ways in which data is most commonly provided, for most deep learning datasets?
- Look up the documentation for
Land try using a few of the new methods that it adds. - Look up the documentation for the Python
pathlibmodule and try using a few methods of thePathclass. - Give two examples of ways that image transformations can degrade the quality of the data.
- What method does fastai provide to view the data in a
DataLoaders? - What method does fastai provide to help you debug a
DataBlock? - Should you hold off on training a model until you have thoroughly cleaned your data?
- What are the two pieces that are combined into cross-entropy loss in PyTorch?
- What are the two properties of activations that softmax ensures? Why is this important?
- When might you want your activations to not have these two properties?
- Calculate the
expandsoftmaxcolumns of <> yourself (i.e., in a spreadsheet, with a calculator, or in a notebook).</li> - Why can't we use
torch.whereto create a loss function for datasets where our label can have more than two categories?- What is the value of log(-2)? Why?
- What are two good rules of thumb for picking a learning rate from the learning rate finder?
- What two steps does the
fine_tunemethod do?- In Jupyter Notebook, how do you get the source code for a method or function?
- What are discriminative learning rates?
- How is a Python
sliceobject interpreted when passed as a learning rate to fastai?- Why is early stopping a poor choice when using 1cycle training?
- What is the difference between
resnet50andresnet101?- What does
</ol> </div> </div> </div>to_fp16do?</div>- Find the paper by Leslie Smith that introduced the learning rate finder, and read it.
- See if you can improve the accuracy of the classifier in this chapter. What's the best accuracy you can achieve? Look on the forums and the book's website to see what other students have achieved with this dataset, and how they did it.
- Why can't we use