def test_produce_when_inference_model_moco():
test_frame = load_frame()
rsp = RemoteSensingPretrainedPrimitive(
hyperparams=rs_hp(
rs_hp.defaults(),
inference_model = 'moco',
use_columns = [1],
),
volumes = {'amdim_weights': amdim_path, 'moco_weights': moco_path},
)
global feature_df
feature_df = rsp.produce(inputs=test_frame).value
_test_output_df(test_frame, feature_df, 2048)
def test_produce_when_inference_model_moco_decompress():
test_frame = load_frame(compress_data = True)
rsp = RemoteSensingPretrainedPrimitive(
hyperparams=rs_hp(
rs_hp.defaults(),
inference_model = 'moco',
use_columns = [1],
decompress_data = True,
),
volumes = {'amdim_weights': amdim_path, 'moco_weights': moco_path},
)
decompress_feature_df = rsp.produce(inputs=test_frame).value
_test_output_df(test_frame, decompress_feature_df, 2048)
assert decompress_feature_df.equals(feature_df)
def test_big_earthnet():
train_inputs, labels = load_big_earthnet()
featurizer = RemoteSensingPretrainedPrimitive(hyperparams=rs_hp(
rs_hp.defaults(),
inference_model='moco',
use_columns=[0],
),
volumes={
'amdim_weights':
amdim_path,
'moco_weights': moco_path
})
features = featurizer.produce(inputs=train_inputs).value
features.to_pickle("dummy.pkl")
# features = pd.read_pickle("dummy.pkl")
iterative_labeling(features, labels)
def load_inputs():
fnames = sorted(glob('/test_data/bigearth-100-single/*/*.tif'))
imnames = sorted(list(set(['_'.join(f.split('_')[:-1]) for f in fnames])))
imgs = [load_patch(img_path).astype(np.float32) for img_path in imnames]
imgs_df = pd.DataFrame({'image_col': imgs, 'dummy_idx': range(len(imgs))})
y = [i.split('/')[3] for i in imnames]
tgts_df = pd.DataFrame({'target': y})
return (d3m_DataFrame(imgs_df), d3m_DataFrame(tgts_df))
train_inputs, labels = load_inputs()
featurizer = RemoteSensingPretrainedPrimitive(hyperparams=rs_hp(
rs_hp.defaults(),
inference_model='moco',
use_columns=[0],
pool_features=False),
volumes={
'amdim_weights': amdim_path,
'moco_weights': moco_path
})
features = featurizer.produce(inputs=train_inputs).value
features = features.drop(columns='dummy_idx')
# features.to_pickle("dummy.pkl")
# labels.to_pickle("labels.pkl")
# features = pd.read_pickle("dummy.pkl")
# labels = pd.read_pickle("labels.pkl")
def test_fit():