def batch_write(self, batch):
fbatch = flatten_batch(batch, nested_sep="/")
# since 'score' is not defined in GenomicRanges, use "."
if os.path.join(self.ranges_key, "score") not in fbatch:
fbatch[os.path.join(self.ranges_key, "score")] = "."
bed_cols = ["chr", "start", "end", "id", "score", "strand"]
cols = [os.path.join(self.ranges_key, x) for x in bed_cols] + \
sorted([x for x in fbatch if x.startswith("preds/")])
df = pd.DataFrame(fbatch)[cols]
df.rename(
columns={os.path.join(self.ranges_key, bc): bc
for bc in bed_cols},
inplace=True)
df.rename(columns={"id": "name"}, inplace=True)
if self.first_pass and self.header:
df.to_csv(self.file_path, sep="\t", index=False)
self.first_pass = False
else:
df.to_csv(self.file_path,
sep="\t",
index=False,
header=None,
mode="a")
def batch_write(self, batch):
df = pd.DataFrame(flatten_batch(batch, nested_sep=self.nested_sep))
df.sort_index(axis=1, inplace=True)
if self.first_pass:
df.to_csv(self.file_path, sep="\t", index=False)
self.first_pass = False
else:
df.to_csv(self.file_path,
sep="\t",
index=False,
header=None,
mode="a")
def batch_write(self, batch):
"""Write a batch of data
# Arguments
batch: batch of data. Either a single `np.array` or a list/dict thereof.
"""
df = pd.DataFrame(flatten_batch(batch, nested_sep=self.nested_sep))
df.sort_index(axis=1, inplace=True)
if self.first_pass:
df.to_csv(self.file_path, sep="\t", index=False)
self.first_pass = False
else:
df.to_csv(self.file_path,
sep="\t",
index=False,
header=None,
mode="a")
def batch_write(self, batch):
"""Write a batch of data
# Arguments
batch: batch of data. Either a single `np.array` or a list/dict thereof.
"""
df = pd.DataFrame(flatten_batch(batch, nested_sep=self.nested_sep))
df.sort_index(axis=1, inplace=True)
if self.first_pass:
self.fp.write(self.file_path,
df,
append=not self.first_pass,
write_index=False,
**self.kwargs)
self.first_pass = False
else:
self.fp.write(self.file_path,
df,
append=not self.first_pass,
write_index=False,
**self.kwargs)
