Multi-label prediction with Planet Amazon dataset

In [ ]:
%reload_ext autoreload
%autoreload 2
%matplotlib inline
In [ ]:
from fastai.vision import *

Getting the data

The planet dataset isn't available on the fastai dataset page due to copyright restrictions. You can download it from Kaggle however. Let's see how to do this by using the Kaggle API as it's going to be pretty useful to you if you want to join a competition or use other Kaggle datasets later on.

First, install the Kaggle API by uncommenting the following line and executing it, or by executing it in your terminal (depending on your platform you may need to modify this slightly to either add source activate fastai or similar, or prefix pip with a path. Have a look at how conda install is called for your platform in the appropriate Returning to work section of https://course.fast.ai/. (Depending on your environment, you may also need to append "--user" to the command.)

In [ ]:
# ! {sys.executable} -m pip install kaggle --upgrade

Then you need to upload your credentials from Kaggle on your instance. Login to kaggle and click on your profile picture on the top left corner, then 'My account'. Scroll down until you find a button named 'Create New API Token' and click on it. This will trigger the download of a file named 'kaggle.json'.

Upload this file to the directory this notebook is running in, by clicking "Upload" on your main Jupyter page, then uncomment and execute the next two commands (or run them in a terminal). For Windows, uncomment the last two commands.

In [ ]:
# ! mkdir -p ~/.kaggle/
# ! mv kaggle.json ~/.kaggle/

# For Windows, uncomment these two commands
# ! mkdir %userprofile%\.kaggle
# ! move kaggle.json %userprofile%\.kaggle

You're all set to download the data from planet competition. You first need to go to its main page and accept its rules, and run the two cells below (uncomment the shell commands to download and unzip the data). If you get a 403 forbidden error it means you haven't accepted the competition rules yet (you have to go to the competition page, click on Rules tab, and then scroll to the bottom to find the accept button).

In [ ]:
path = Config.data_path()/'planet'
path.mkdir(parents=True, exist_ok=True)
path
Out[ ]:
PosixPath('/home/ubuntu/.fastai/data/planet')
In [ ]:
# ! kaggle competitions download -c planet-understanding-the-amazon-from-space -f train-jpg.tar.7z -p {path}  
# ! kaggle competitions download -c planet-understanding-the-amazon-from-space -f train_v2.csv -p {path}  
# ! unzip -q -n {path}/train_v2.csv.zip -d {path}

To extract the content of this file, we'll need 7zip, so uncomment the following line if you need to install it (or run sudo apt install p7zip-full in your terminal).

In [ ]:
# ! conda install --yes --prefix {sys.prefix} -c haasad eidl7zip

And now we can unpack the data (uncomment to run - this might take a few minutes to complete).

In [ ]:
# ! 7za -bd -y -so x {path}/train-jpg.tar.7z | tar xf - -C {path.as_posix()}

Multiclassification

Contrary to the pets dataset studied in last lesson, here each picture can have multiple labels. If we take a look at the csv file containing the labels (in 'train_v2.csv' here) we see that each 'image_name' is associated to several tags separated by spaces.

In [ ]:
df = pd.read_csv(path/'train_v2.csv')
df.head()
Out[ ]:
image_name tags
0 train_0 haze primary
1 train_1 agriculture clear primary water
2 train_2 clear primary
3 train_3 clear primary
4 train_4 agriculture clear habitation primary road

To put this in a DataBunch while using the data block API, we then need to using ImageList (and not ImageDataBunch). This will make sure the model created has the proper loss function to deal with the multiple classes.

In [ ]:
tfms = get_transforms(flip_vert=True, max_lighting=0.1, max_zoom=1.05, max_warp=0.)

We use parentheses around the data block pipeline below, so that we can use a multiline statement without needing to add '\'.

In [ ]:
np.random.seed(42)
src = (ImageList.from_csv(path, 'train_v2.csv', folder='train-jpg', suffix='.jpg')
       .split_by_rand_pct(0.2)
       .label_from_df(label_delim=' '))
In [ ]:
data = (src.transform(tfms, size=128)
        .databunch().normalize(imagenet_stats))

show_batch still works, and show us the different labels separated by ;.

In [ ]:
data.show_batch(rows=3, figsize=(12,9))

To create a Learner we use the same function as in lesson 1. Our base architecture is resnet50 again, but the metrics are a little bit differeent: we use accuracy_thresh instead of accuracy. In lesson 1, we determined the predicition for a given class by picking the final activation that was the biggest, but here, each activation can be 0. or 1. accuracy_thresh selects the ones that are above a certain threshold (0.5 by default) and compares them to the ground truth.

As for Fbeta, it's the metric that was used by Kaggle on this competition. See here for more details.

In [ ]:
arch = models.resnet50
In [ ]:
acc_02 = partial(accuracy_thresh, thresh=0.2)
f_score = partial(fbeta, thresh=0.2)
learn = cnn_learner(data, arch, metrics=[acc_02, f_score])
Downloading: "https://download.pytorch.org/models/resnet50-19c8e357.pth" to /home/ubuntu/.torch/models/resnet50-19c8e357.pth
100%|██████████| 102502400/102502400 [00:01<00:00, 100859665.66it/s]

We use the LR Finder to pick a good learning rate.

In [ ]:
learn.lr_find()
LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.
In [ ]:
learn.recorder.plot()

Then we can fit the head of our network.

In [ ]:
lr = 0.01
In [ ]:
learn.fit_one_cycle(5, slice(lr))
Total time: 03:30

epoch train_loss valid_loss accuracy_thresh fbeta
1 0.125128 0.110038 0.944010 0.904893
2 0.112533 0.101303 0.955964 0.910774
3 0.100574 0.093255 0.955187 0.918653
4 0.096056 0.087997 0.954983 0.924016
5 0.092320 0.086761 0.956400 0.925110
In [ ]:
learn.save('stage-1-rn50')

...And fine-tune the whole model:

In [ ]:
learn.unfreeze()
In [ ]:
learn.lr_find()
learn.recorder.plot()
LR Finder complete, type {learner_name}.recorder.plot() to see the graph.
In [ ]:
learn.fit_one_cycle(5, slice(1e-5, lr/5))
Total time: 04:00

epoch train_loss valid_loss accuracy_thresh fbeta
1 0.097016 0.094868 0.952004 0.916215
2 0.095774 0.088899 0.954540 0.922340
3 0.090646 0.085958 0.959249 0.924921
4 0.085097 0.083291 0.958849 0.928195
5 0.079197 0.082855 0.958602 0.928259
In [ ]:
learn.save('stage-2-rn50')
In [ ]:
data = (src.transform(tfms, size=256)
        .databunch().normalize(imagenet_stats))

learn.data = data
data.train_ds[0][0].shape
Out[ ]:
torch.Size([3, 256, 256])
In [ ]:
learn.freeze()
In [ ]:
learn.lr_find()
learn.recorder.plot()
LR Finder complete, type {learner_name}.recorder.plot() to see the graph.
In [ ]:
lr=1e-2/2
In [ ]:
learn.fit_one_cycle(5, slice(lr))
Total time: 09:01

epoch train_loss valid_loss accuracy_thresh fbeta
1 0.087761 0.085013 0.958006 0.926066
2 0.087641 0.083732 0.958260 0.927459
3 0.084250 0.082856 0.958485 0.928200
4 0.082347 0.081470 0.960091 0.929166
5 0.078463 0.080984 0.959249 0.930089
In [ ]:
learn.save('stage-1-256-rn50')
In [ ]:
learn.unfreeze()
In [ ]:
learn.fit_one_cycle(5, slice(1e-5, lr/5))
Total time: 11:25

epoch train_loss valid_loss accuracy_thresh fbeta
1 0.082938 0.083548 0.957846 0.927756
2 0.086312 0.084802 0.958718 0.925416
3 0.084824 0.082339 0.959975 0.930054
4 0.078784 0.081425 0.959983 0.929634
5 0.074530 0.080791 0.960426 0.931257
In [ ]:
learn.recorder.plot_losses()
In [ ]:
learn.save('stage-2-256-rn50')

You won't really know how you're going until you submit to Kaggle, since the leaderboard isn't using the same subset as we have for training. But as a guide, 50th place (out of 938 teams) on the private leaderboard was a score of 0.930.

In [ ]:
learn.export()

fin

(This section will be covered in part 2 - please don't ask about it just yet! :) )

In [ ]:
#! kaggle competitions download -c planet-understanding-the-amazon-from-space -f test-jpg.tar.7z -p {path}  
#! 7za -bd -y -so x {path}/test-jpg.tar.7z | tar xf - -C {path}
#! kaggle competitions download -c planet-understanding-the-amazon-from-space -f test-jpg-additional.tar.7z -p {path}  
#! 7za -bd -y -so x {path}/test-jpg-additional.tar.7z | tar xf - -C {path}
In [ ]:
test = ImageList.from_folder(path/'test-jpg').add(ImageList.from_folder(path/'test-jpg-additional'))
len(test)
Out[ ]:
61191
In [ ]:
learn = load_learner(path, test=test)
preds, _ = learn.get_preds(ds_type=DatasetType.Test)
In [ ]:
thresh = 0.2
labelled_preds = [' '.join([learn.data.classes[i] for i,p in enumerate(pred) if p > thresh]) for pred in preds]
In [ ]:
labelled_preds[:5]
Out[ ]:
['agriculture cultivation partly_cloudy primary road',
 'clear haze primary water',
 'agriculture clear cultivation primary',
 'clear primary',
 'partly_cloudy primary']
In [ ]:
fnames = [f.name[:-4] for f in learn.data.test_ds.items]
In [ ]:
df = pd.DataFrame({'image_name':fnames, 'tags':labelled_preds}, columns=['image_name', 'tags'])
In [ ]:
df.to_csv(path/'submission.csv', index=False)
In [ ]:
! kaggle competitions submit planet-understanding-the-amazon-from-space -f {path/'submission.csv'} -m "My submission"
Warning: Your Kaggle API key is readable by other users on this system! To fix this, you can run 'chmod 600 /home/ubuntu/.kaggle/kaggle.json'
100%|██████████████████████████████████████| 2.18M/2.18M [00:02<00:00, 1.05MB/s]
Successfully submitted to Planet: Understanding the Amazon from Space

Private Leaderboard score: 0.9296 (around 80th)