mirror of
https://github.com/allaunthefox/Research-Stack.git
synced 2026-07-31 03:05:21 +00:00
9 lines
185 B
TOML
9 lines
185 B
TOML
limit_warc_files = 5
|
|
graph_base_path = "data/webgraph"
|
|
batch_size = 1
|
|
canonical_index_path = "data/canonical"
|
|
|
|
[warc_source]
|
|
folder = "./data"
|
|
names = ["sample.warc.gz"]
|
|
type = "Local"
|