def get(self) -> pd.DataFrame:
ns = self.__namespace or '_'
ds = self.__dataset
version = self.__version or 'latest'
resp = client.get('/datasets/' + ns + '/' + ds + '/versions/' + version + '/data')
return pandavro.from_avro(BytesIO(resp.content))
def test_buffer_e2e(dataframe):
tf = NamedTemporaryFile()
pdx.to_avro(tf.name, dataframe)
with open(tf.name, 'rb') as f:
expect = pdx.from_avro(BytesIO(f.read()))
assert_frame_equal(expect, dataframe)
f.close()
def test_delegation(dataframe):
tf = NamedTemporaryFile()
pdx.to_avro(tf.name, dataframe)
expect = pdx.from_avro(tf.name)
expect['DateTime64'] = expect['DateTime64'].astype(
np.dtype('datetime64[ns]'))
assert_frame_equal(expect, dataframe)
def test_append(dataframe):
tf = NamedTemporaryFile()
pdx.to_avro(tf.name, dataframe[0:int(dataframe.shape[0] / 2)])
pdx.to_avro(tf.name, dataframe[int(dataframe.shape[0] / 2):], append=True)
expect = pdx.from_avro(tf.name)
expect['DateTime64'] = expect['DateTime64'].astype(
np.dtype('datetime64[ns]'))
assert_frame_equal(expect, dataframe)
def _create_tfrecord(file):
data = pdx.from_avro("TrainTemp/" + file)
fname = file.replace("avro", "tfrecords")
with tf.python_io.TFRecordWriter("TrainData/" + fname) as writer:
for _, row in data.iterrows():
index = [idx[row[col]] for col, idx in COLUMNS.items()]
value = np.full(len(index), 1)
label = row.click
example = tf.train.Example()
example.features.feature["index"].int64_list.value.extend(index)
example.features.feature["value"].int64_list.value.extend(value)
example.features.feature["label"].int64_list.value.append(label)
writer.write(example.SerializeToString())
def test_delegation(dataframe):
tf = NamedTemporaryFile()
pdx.to_avro(tf.name, dataframe)
expect = pdx.from_avro(tf.name)
assert_frame_equal(expect, dataframe)
def test_file_path_e2e(dataframe):
tf = NamedTemporaryFile()
pdx.to_avro(tf.name, dataframe)
expect = pdx.from_avro(tf.name)
assert_frame_equal(expect, dataframe)
def main():
weather = pdx.from_avro('weather.avro')
print(weather)
pdx.to_avro('weather_out.avro', weather)
def _load_avro(dirs):
gcs = gcsio.GcsIO()
with gcs.open(dirs, 'rb') as f:
return pdx.from_avro(f)
def deserialize_avro_str_to_pandas(avro_str: str, schema: dict = None) -> pd.DataFrame:
return pandavro.from_avro(io.BytesIO(base64.b64decode(avro_str)), schema)
{
"name": "유재석",
"birth": "1972-08-14",
"job": "MC, 개그맨"
},
{
"name": "강호동",
"birth": "1970-05-11",
"job": "MC, 개그맨"
},
{
"name": "김구라",
"birth": "1970-10-03",
"job": "MC, 개그맨"
},
]
print(type(member))
print()
# DataFrame
df1 = pd.DataFrame.from_records(member)
print(df1)
print()
# Avro 쓰기
pandavro.to_avro("./data/csv/member.avro", df1)
# Avro 읽기
df2 = pandavro.from_avro("./data/csv/member.avro")
print(df2)
