Files
2026-03-14 21:48:05 +00:00

56 lines
2.1 KiB
YAML

temporary-directory: null # Directory for local disk like /tmp, /scratch, or /local
visualization:
engine: null # Default visualization engine to use when calling `.visualize()` on a collection
tokenize:
ensure-deterministic: false # If true, tokenize will error instead of falling back to uuids
dataframe:
backend: "pandas" # Backend dataframe library for input IO and data creation
shuffle:
method: null
compression: null # compression for on disk-shuffling. Partd supports ZLib, BZ2, SNAPPY
parquet:
metadata-task-size-local: 512 # Number of files per local metadata-processing task
metadata-task-size-remote: 1 # Number of files per remote metadata-processing task
minimum-partition-size: 75000000
convert-string: null # Whether to convert string-like data to pyarrow strings
query-planning: null # Whether to use dask-expr
array:
backend: "numpy" # Backend array library for input IO and data creation
chunk-size: "128MiB"
chunk-size-tolerance: 1.25 # Tolerance for different algorithms when creating output chunks.
rechunk:
method: null # Rechunking method to use
threshold: 4
svg:
size: 120 # pixels
slicing:
split-large-chunks: null # How to handle large output chunks in slicing. Warns by default.
query-planning: null # Whether to use dask-expr
optimization:
annotations:
fuse: true # Automatically fuse compatible annotations on layers
fuse:
active: null # Treat as false for dask.dataframe, true for everything else
ave-width: 1
max-width: null # 1.5 + ave_width * log(ave_width + 1)
max-height: .inf
max-depth-new-edges: null # ave_width * 1.5
subgraphs: null # true for dask.dataframe, false for everything else
rename-keys: true
admin:
traceback:
shorten:
- concurrent[\\\/]futures[\\\/]
- dask[\\\/](base|core|local|multiprocessing|optimization|threaded|utils)\.py
- dask[\\\/]array[\\\/]core\.py
- dask[\\\/]dataframe[\\\/](core|methods)\.py
- distributed[\\\/](client|scheduler|utils|worker)\.py
- tornado[\\\/]gen\.py
- pandas[\\\/]core[\\\/]