CarlinLiao
commented
1 year ago Added a test, so this will be good to merge once we can get it to pass.
The strange thing is, during the nextflow run, it errors and complains that torch is missing, even though the workflow Docker image is now built off of a torch image. Do the nextflow instances not run on the workflow Docker image? The other nextflow configurations don't need to do any special installations!
┃ cggnn ┈┈┈
(Started recording Docker compose logs for this test.)
process.executor = 'local'
env.workflow_ = 'cggnn'
env.db_config_file_ = '../db/.spt_db.config.container'
env.study_name_ = 'Melanoma intralesional IL2'
env.strata_ = '[5 7]'
env.validation_data_percent_ = '15'
env.test_data_percent_ = '0'
env.disable_channels_ = 'false'
env.disable_phenotypes_ = 'false'
env.cells_per_slide_target_ = '5000'
env.target_name_ = 'P Tumor'
env.in_ram_ = 'true'
env.batch_size_ = '1'
env.epochs_ = '5'
env.learning_rate_ = '1e-3'
env.k_folds_ = '0'
env.explainer_model_ = 'pp'
env.merge_rois_ = 'true'
env.prune_misclassified_ = 'false'
env.output_prefix_ = 'mi2'
env.upload_importances_ = 'false'N E X T F L O W ~ version 23.10.0
Launching `./main.nf` [fervent_galileo] DSL2 - revision: fd0ce3c646
executor > local (2)
[07/45f395] process > echo_environment_variables [100%] 1 of 1 ✔
[b6/ea7bff] process > run_cggnn [ 0%] 0 of 1
ERROR ~ Error executing process > 'run_cggnn'
Caused by:
Process `run_cggnn` terminated with an error exit status (1)
Command executed:
spt cggnn run --spt_db_config_location=".spt_db.config.container" --study="Melanoma intralesional IL2" --strata=[5 7] --validation-data-percent=15 --test-data-percent=0 $(if [[ "false" = true ]]; then echo "--disable-channels"; fi) $(if [[ "false" = true ]]; then echo "--disable_phenotypes"; fi) --cells-per-slide-target=5000 --target-name="P Tumor" $(if [[ "true" = true ]]; then echo "--in_ram"; fi) --batch-size=1 --epochs=5 --learning-rate=1e-3 --k-folds=0 --explainer-model="pp" $(if [[ "true" = true ]]; then echo "--merge_rois"; fi) $(if [[ "false" = true ]]; then echo "--prune_misclassified"; fi) --output-prefix="mi2" $(if [[ "false" = true ]]; then echo "--upload_importances"; fi)
Command exit status:
1
Command output:
Got a module not found error.
Did you install the required extras with:
pip install "spatialprofilingtoolbox[cggnn]"
?
Command error:
Got a module not found error.
Did you install the required extras with:
pip install "spatialprofilingtoolbox[cggnn]"
?
Traceback (most recent call last):
File "/usr/local/lib/python3.11/dist-packages/spatialprofilingtoolbox/cggnn/scripts/run.py", line 15, in <module>
SuggestExtrasException(e, 'cggnn')
File "/usr/local/lib/python3.11/dist-packages/spatialprofilingtoolbox/standalone_utilities/module_load_error.py", line 7, in __init__
executor > local (2)
[07/45f395] process > echo_environment_variables [100%] 1 of 1 ✔
[b6/ea7bff] process > run_cggnn [100%] 1 of 1, failed: 1 ✘
ERROR ~ Error executing process > 'run_cggnn'
Caused by:
Process `run_cggnn` terminated with an error exit status (1)
Command executed:
spt cggnn run --spt_db_config_location=".spt_db.config.container" --study="Melanoma intralesional IL2" --strata=[5 7] --validation-data-percent=15 --test-data-percent=0 $(if [[ "false" = true ]]; then echo "--disable-channels"; fi) $(if [[ "false" = true ]]; then echo "--disable_phenotypes"; fi) --cells-per-slide-target=5000 --target-name="P Tumor" $(if [[ "true" = true ]]; then echo "--in_ram"; fi) --batch-size=1 --epochs=5 --learning-rate=1e-3 --k-folds=0 --explainer-model="pp" $(if [[ "true" = true ]]; then echo "--merge_rois"; fi) $(if [[ "false" = true ]]; then echo "--prune_misclassified"; fi) --output-prefix="mi2" $(if [[ "false" = true ]]; then echo "--upload_importances"; fi)
Command exit status:
1
Command output:
Got a module not found error.
Did you install the required extras with:
pip install "spatialprofilingtoolbox[cggnn]"
?
Command error:
Got a module not found error.
Did you install the required extras with:
pip install "spatialprofilingtoolbox[cggnn]"
?
Traceback (most recent call last):
File "/usr/local/lib/python3.11/dist-packages/spatialprofilingtoolbox/cggnn/scripts/run.py", line 15, in <module>
SuggestExtrasException(e, 'cggnn')
File "/usr/local/lib/python3.11/dist-packages/spatialprofilingtoolbox/standalone_utilities/module_load_error.py", line 7, in __init__
raise module_not_found_error
File "/usr/local/lib/python3.11/dist-packages/spatialprofilingtoolbox/cggnn/scripts/run.py", line 12, in <module>
from cggnn.run import run
File "/usr/local/lib/python3.11/dist-packages/cggnn/__init__.py", line 3, in <module>
from cggnn.generate_graphs import generate_graphs
File "/usr/local/lib/python3.11/dist-packages/cggnn/generate_graphs.py", line 9, in <module>
from torch import Tensor, FloatTensor, IntTensor # pylint: disable=no-name-in-module
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ModuleNotFoundError: No module named 'torch'
Work dir:
/mount_sources/workflow/work/b6/ea7bffd8f25b843fd0685b32209536
Tip: when you have fixed the problem you can continue the execution adding the option `-resume` to the run command line
-- Check '.nextflow.log' file for details
head: cannot open 'results/mi2_importances.csv' for reading: No such file or directory
Error during nextflow run.
Docker compose logs:
no such service: testing-api-server
no such service: testing-fast-counts-server
┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈┈ Failed. (48s) 
jimmymathews
Re-pursuant to #85, these changes add a cg-gnn Nextflow configuration to the list of workflows.