Closed kevinkle closed 5 years ago
Merging #37 into master will decrease coverage by
1.76%
. The diff coverage is84.9%
.
@@ Coverage Diff @@
## master #37 +/- ##
==========================================
- Coverage 93.46% 91.69% -1.77%
==========================================
Files 11 10 -1
Lines 367 337 -30
==========================================
- Hits 343 309 -34
- Misses 24 28 +4
Impacted Files | Coverage Δ | |
---|---|---|
prairiedog/prairiedog.py | 66.66% <0%> (-8.34%) |
:arrow_down: |
prairiedog/kmers.py | 94.28% <100%> (+0.08%) |
:arrow_up: |
prairiedog/subgraph_ref.py | 100% <100%> (ø) |
:arrow_up: |
prairiedog/graph_ref.py | 94.01% <83.33%> (-5.99%) |
:arrow_down: |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update c52edea...eff38f8. Read the comment docs.
Instead of counting, just set a very high max n
Init the GraphRef during KmerGraphinit
Fix: use class attrs for now instead of config
Read K from config if not supplied
Read K from config if not supplied
Remove the assert len(lines) == gr.n check since we now use max_n
Remove the assert len(lines) == gr.n check since we now use max_n
Check datrie and Py3.7 support
Install datrie via its github repo instead of PyPI
Separate tox rule for [snakemake]
Separate tox rule for [snakemake]
Separate tox rule for [snakemake]
Initial draft of Snakefile
Fix Snakefile
Debug why Snakemake isnt called in tox
Fix log call in KmerGraph
Delete KmerGraph work and solely use Snakemake to drive dataset generation
Use protocol=4 for pickle to bypass the hardcoded 4GB pickle limit (PEP3154)
Use Snakemake config to set node_labels csv
Use Snakemake config to set node_labels csv
Use Snakemake config to set node_labels csv
Check what GraphRef.close() is writing files out as
Try using numpy.loadtxt() instead of parsing
Switch Snakefile to use dill instead of celery
Load previous files via pickle for now
Fixed: wasnt dill, but the config was using MICs from test
Fixed: wasnt dill, but the config was using MICs from test
Add dill to requirements
Dont init the numpy arrays until subgraphs are created
Dont init the numpy arrays until subgraphs are created
Init node arrays on first call instead of via append()
Still dump the GraphRef instance in Snakefile
Dump the appended GraphRef to a different file
Add a sanity check before saving NumPy arrays
Snakemake is calling append with the required subgraphs, but not sure why the arrays are zero
Snakemake is calling append with the required subgraphs, but not sure why the arrays are zero
Use outputs/graphref_final.pkl
Call snakemake clean between tests
deleted: prairiedog/kmer_graph.py