File size: 404 Bytes
fb12e65
 
eef88e0
fb12e65
 
 
72d08a5
 
fb12e65
c624b60
fb12e65
1
2
3
4
5
6
7
8
9
10
11
import pandas as pd

_SPLIT = "train"

df = pd.read_parquet("data/dataset_audio_" + _SPLIT +".parquet.gzip")

clipped_df = df.filter(["label_string", "probability", "probability_vector", "prediction",
                "prediction_string", "embedding_reduced"], axis=1)

clipped_df = clipped_df.sort_values(by=["label_string"])
clipped_df.to_parquet("data/dataset_audio_" + _SPLIT +"_clipped.parquet.gzip")