def load_model(filesystem: S3FS, config: PredictionConfig) -> ResUnetA:
""" Copy the model locally if not existing and load it """
if not os.path.exists(f'{config.temp_model_path}/{config.model_name}'):
if not filesystem.exists(
f'{config.model_path}/{config.model_name}/checkpoints/'):
filesystem.makedirs(
f'{config.model_path}/{config.model_name}/checkpoints/')
copy_dir(filesystem,
f'{config.model_path}/{config.model_name}/checkpoints/',
f'{config.temp_model_path}/{config.model_name}',
'checkpoints')
copy_file(filesystem,
f'{config.model_path}/{config.model_name}/model_cfg.json',
f'{config.temp_model_path}/{config.model_name}',
'model_cfg.json')
input_shape = dict(
features=[None, config.height, config.width, config.n_channels])
with open(f'{config.temp_model_path}/{config.model_name}/model_cfg.json',
'r') as jfile:
model_cfg = json.load(jfile)
# initialise model from config, build, compile and load trained weights
model = ResUnetA(model_cfg)
model.build(input_shape)
model.net.compile()
model.net.load_weights(
f'{config.temp_model_path}/{config.model_name}/checkpoints/model.ckpt')
return model
def load_metadata(filesystem: S3FS, config: PredictionConfig) -> pd.DataFrame:
""" Load DataFrame with info about normalisation factors """
metadata_dir = os.path.dirname(config.metadata_path)
if not filesystem.exists(metadata_dir):
filesystem.makedirs(metadata_dir)
df = pd.read_csv(filesystem.open(f'{config.metadata_path}'))
normalisation_factors = df.groupby(
pd.to_datetime(df.timestamp).dt.to_period("M")).max()
normalisation_factors['month'] = pd.to_datetime(
normalisation_factors.timestamp).dt.month
return normalisation_factors
def npz_dir_dataset(file_dir_or_list: Union[str, List[str]],
features: dict,
metadata_path: str,
fold: int = None,
randomize: bool = True,
num_parallel: int = 5,
shuffle_size: int = 500,
filesystem: S3FS = None,
npz_from_s3: bool = False) -> tf.data.Dataset:
""" Creates a tf.data.Dataset from a directory containing numpy .npz files.
Files are loaded lazily when needed. `num_parallel` files are read in parallel and interleaved together.
:param file_dir_or_list: directory containing .npz files or a list of paths to .npz files
:param features: dict of (`field` -> `feature_name`) mappings, where `field` is the field in the .npz array
and `feature_name` is the name of the feature it is saved to.
:param fold: in k-fold validation, fold to consider when querying the patchlet info dataframe
:param randomize: whether to shuffle the samples of the dataset or not, defaults to `True`
:param num_parallel: number of files to read in parallel and intereleave, defaults to 5
:param shuffle_size: buffer size for shuffling file order, defaults to 500
:param metadata_path: path to input csv files with patchlet information
:param filesystem: filesystem to access bucket, defaults to None
:param npz_from_s3: if True, npz files are loaded from S3 bucket, otherwise from local disk
:return: dataset containing examples merged from files
"""
files = file_dir_or_list
if npz_from_s3:
assert filesystem is not None
# If dir, then list files
if isinstance(file_dir_or_list, str):
if filesystem and not filesystem.isdir(file_dir_or_list):
filesystem.makedirs(file_dir_or_list)
dir_list = os.listdir(
file_dir_or_list) if not npz_from_s3 else filesystem.listdir(
file_dir_or_list)
files = [os.path.join(file_dir_or_list, f) for f in dir_list]
fields = list(features.keys())
# Read one file for shape info
file = next(iter(files))
data = np.load(file) if not npz_from_s3 else np.load(
filesystem.openbin(file))
np_arrays = [data[f] for f in fields]
# Append norm arrays
perc99, meanstd_mean, meanstd_median, meanstd_std = _construct_norm_arrays(
file, metadata_path, fold, filesystem)
np_arrays.append(perc99)
np_arrays.append(meanstd_mean)
np_arrays.append(meanstd_median)
np_arrays.append(meanstd_std)
# Read shape and type info
# types = tuple(arr.dtype for arr in np_arrays)
types = (tf.uint16, tf.float32, tf.float32, tf.float32, tf.float64,
tf.float64, tf.float64, tf.float64)
shapes = tuple(arr.shape[1:] for arr in np_arrays)
# Create datasets
datasets = [
_npz_file_lazy_dataset(file,
fields,
types,
shapes,
metadata_path,
fold=fold,
filesystem=filesystem,
npz_from_s3=npz_from_s3) for file in files
]
ds = tf.data.Dataset.from_tensor_slices(datasets)
# Shuffle files and interleave multiple files in parallel
if randomize:
ds = ds.shuffle(shuffle_size)
ds = ds.interleave(lambda x: x, cycle_length=num_parallel)
return ds
