search

Flipajs / FERDA

MIT License
1 stars 0 forks source link

export failure #5

Open bcasillas opened 7 years ago

bcasillas commented 7 years ago

commit 1e8103e with error: core/region/region_manager.py:31: UserWarning: cache size limit -1 - infinity is not supported right now!!! warnings.warn("cache size limit -1 - infinity is not supported right now!!!") Traceback (most recent call last): File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main "main", fname, loader, pkg_name) File "/usr/lib/python2.7/runpy.py", line 72, in _run_code exec code in run_globals File "/nfs/m8/home/casillas/ferda/FERDA/scripts/export/export_part.py", line 195, in assembly_after_parallelization(bgcomp) File "core/bg_computer_assembling.py", line 73, in assembly_after_parallelization mergeparts(bgcomp.project.gm, g, relevant_vertices, bgcomp.project, rmold, chm) File "core/bg_computer_assembling.py", line 290, in merge_parts new_rm.add(old_reg) File "core/region/region_manager.py", line 120, in add self.add_tocache(id, regions) File "core/region/region_manager.py", line 138, in add_tocache self.cache[id] = region File "libs/cachetools/lru.py", line 21, in setitem cache_setitem(self, key, value) File "libs/cachetools/cache.py", line 50, in setitem raise ValueError('value too large') ValueError: value too large

Flipajs commented 7 years ago

should be fixed in cd8a1f1

(added estimate for cache size in bg_computer to allow in place assembly without database. Cachetools doesn't support infinite cache size...)

could you please try it and leave me a message about result?

bcasillas commented 7 years ago

commit cd8a1f1

Traceback (most recent call last): File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main "main", fname, loader, pkg_name) File "/usr/lib/python2.7/runpy.py", line 72, in _run_code exec code in run_globals File "/nfs/m8/home/casillas/ferda/FERDA/scripts/export/export_part.py", line 195, in assembly_after_parallelization(bgcomp) File "core/bg_computer_assembling.py", line 27, in assembly_after_parallelization cache_size_limit = int(q len(bgcomp.project.animals) bgcomp.frames_in_row_last * bgcomp.part_num) AttributeError: FakeBGComp instance has no attribute 'frames_in_row_last'

bcasillas commented 7 years ago

Traceback (most recent call last): File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main "main", fname, loader, pkg_name) File "/usr/lib/python2.7/runpy.py", line 72, in _run_code exec code in run_globals File "/nfs/m8/home/casillas/ferda/FERDA/scripts/export/export_part.py", line 196, in assembly_after_parallelization(bgcomp) File "core/bg_computer_assembling.py", line 126, in assembly_after_parallelization from scripts.regions_stats import learn_assignments, add_score_to_edges, tracklet_stats File "scripts/regions_stats.py", line 8, in from skimage.data import binary_blobs ImportError: cannot import name binary_blobs

bcasillas commented 7 years ago

commit: bc904ce can you include module in FERDA? or should we install in all nodes?

Traceback (most recent call last): File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main "main", fname, loader, pkg_name) File "/usr/lib/python2.7/runpy.py", line 72, in _run_code exec code in run_globals File "/nfs/m8/home/casillas/ferda/FERDA/scripts/export/export_part.py", line 196, in assembly_after_parallelization(bgcomp) File "core/bg_computer_assembling.py", line 126, in assembly_after_parallelization from scripts.regions_stats import learn_assignments, add_score_to_edges, tracklet_stats File "scripts/regions_stats.py", line 5, in import hickle ImportError: No module named hickle

Flipajs commented 7 years ago

If there is h5py installed on server (easy check will be run python and try import h5py). I can include library into repository, if not, we need to install this (or I can change to pickle, but for these cases hickle is much much faster).

bcasillas commented 7 years ago

h5py seems to be installed

Flipajs commented 7 years ago

ok then... Commit d877c9b should work for you.

bcasillas commented 7 years ago

still couldn't import hickle

bcasillas commented 7 years ago

Traceback (most recent call last): File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main "main", fname, loader, pkg_name) File "/usr/lib/python2.7/runpy.py", line 72, in _run_code exec code in run_globals File "/nfs/m8/home/casillas/ferda/FERDA/scripts/export/export_part.py", line 196, in assembly_after_parallelization(bgcomp) File "core/bg_computer_assembling.py", line 126, in assembly_after_parallelization from scripts.regions_stats import learn_assignments, add_score_to_edges, tracklet_stats File "scripts/regions_stats.py", line 11, in from sklearn.ensemble import RandomForestClassifier, IsolationForest ImportError: cannot import name IsolationForest

bcasillas commented 7 years ago

sklearn version 0.14.1