|
|
import os |
|
|
import pyarrow.parquet as pq |
|
|
import datasets |
|
|
import pandas as pd |
|
|
import numpy as np |
|
|
import io |
|
|
|
|
|
class WyvernEOConfig(datasets.BuilderConfig): |
|
|
def __init__(self, **kwargs): |
|
|
super().__init__(**kwargs) |
|
|
|
|
|
class WyvernEO(datasets.GeneratorBasedBuilder): |
|
|
BUILDER_CONFIGS = [ |
|
|
WyvernEOConfig(name="default", version=datasets.Version("1.0.0"), description="Wyvern EO imagery with masks") |
|
|
] |
|
|
|
|
|
DEFAULT_CONFIG_NAME = "default" |
|
|
|
|
|
def _info(self): |
|
|
return datasets.DatasetInfo( |
|
|
description="Wyvern dataset at 2400m tiles with imagery and quality masks.", |
|
|
features=datasets.Features({ |
|
|
"tile_id": datasets.Value("string"), |
|
|
"scene_id": datasets.Value("string"), |
|
|
"start_datetime": datasets.Value("timestamp[us]"), |
|
|
"end_datetime": datasets.Value("timestamp[us]"), |
|
|
"image": datasets.Value("binary"), |
|
|
"data_mask": datasets.Value("binary"), |
|
|
"pixel_quality_mask": datasets.Value("binary"), |
|
|
"shape": [datasets.Value("int32")], |
|
|
}), |
|
|
supervised_keys=None, |
|
|
) |
|
|
|
|
|
def _split_generators(self, dl_manager): |
|
|
|
|
|
parquet_dir = "datasets_EO/Wyvern/grid_2400m" |
|
|
return [ |
|
|
datasets.SplitGenerator( |
|
|
name=datasets.Split.TRAIN, |
|
|
gen_kwargs={"folder_path": parquet_dir}, |
|
|
) |
|
|
] |
|
|
|
|
|
def _generate_examples(self, folder_path): |
|
|
idx = 0 |
|
|
for fname in sorted(os.listdir(folder_path)): |
|
|
if fname.endswith(".parquet"): |
|
|
fpath = os.path.join(folder_path, fname) |
|
|
table = pq.read_table(fpath) |
|
|
df = table.to_pandas() |
|
|
for _, row in df.iterrows(): |
|
|
yield idx, { |
|
|
"tile_id": row["tile_id"], |
|
|
"scene_id": row["scene_id"], |
|
|
"start_datetime": row["start_datetime"], |
|
|
"end_datetime": row["end_datetime"], |
|
|
"image": row["image"], |
|
|
"data_mask": row["data_mask"], |
|
|
"pixel_quality_mask": row["pixel_quality_mask"], |
|
|
"shape": row["shape"], |
|
|
} |
|
|
idx += 1 |
|
|
|
