ca10c185d94470054d693f19e7691523dbe3ec55,skll/data/readers.py,NDJReader,_sub_read,#NDJReader#Any#,434

Before Change

            line = line.strip()

            // If this is a comment line or a blank line, move on
            if line.startswith("//") or not line:
                continue

            // Process good lines
            example = json.loads(line)
            // Convert all IDs to strings initially,
            // for consistency with csv and megam formats.
            curr_id = str(example.get("id",
                                      "EXAMPLE_{}".format(example_num)))
            class_name = (safe_float(example["y"],
                                     replace_dict=self.class_map)
                          if "y" in example else None)
            example = example["x"]

            if self.ids_to_floats:
                try:
                    curr_id = float(curr_id)
                except ValueError:
                    raise ValueError(("You set ids_to_floats to true, but" +
                                      " ID {} could not be converted to " +
                                      "float").format(curr_id))

After Change

        features : list of dicts
            The features for the features set.
        
        with open(f, "r" if PY3 else "rb") as buff:
            lines = [json.loads(line.strip()) for line in buff
                     if line.strip() and not line.startswith("//")]

        // create a data frame; if it"s empty,
        // then return `_parse_dataframe()`, which
        // will raise an error
        df = pd.DataFrame(lines)
        if df.empty:
            return self._parse_dataframe(df, None, None)

Instances