def extract_features_parallel(feature_certificate,
image_certificate,
model_certificate,
feature_config,
feature_hash,
convolve_func_name,
batch_size):
conn = pm.Connection(document_class = SON)
db = conn[DB_NAME]
image_certdict = cPickle.load(open(image_certificate))
image_hash = image_certdict['image_hash']
image_args = image_certdict['args']
model_certdict = cPickle.load(open(model_certificate))
model_hash = model_certdict['model_hash']
model_args = model_certdict['args']
conn = pm.Connection(document_class = SON)
db = conn[DB_NAME]
f_coll = db['features.files']
f_fs = gridfs.GridFS(db,'features')
remove_existing(f_coll,f_fs,feature_hash)
limits = get_feature_batches(image_hash,model_hash,db['images.files'],db['models.files'],batch_size = batch_size)
jobids = []
if convolve_func_name == 'numpy':
queueName = 'extraction_cpu.q'
elif convolve_func_name == 'pyfft':
queueName = 'extraction_gpu.q'
opstring = '-l qname=' + queueName + ' -o /home/render -e /home/render'
for (ind,limit) in enumerate(limits):
im_from,im_to,m_from,m_to = limit
jobid = qsub(extract_features_core,[(image_certificate,
model_certificate,
feature_hash,
image_hash,
model_hash,
convolve_func_name),
{'im_skip':im_from,
'im_limit':im_to-im_from,
'm_skip':m_from,
'm_limit':m_to-m_from
}],
opstring=opstring)
jobids.append(jobid)
createCertificateDict(feature_certificate,{'feature_hash':feature_hash,
'image_hash':image_hash,
'model_hash':model_hash,
'args':feature_config})
return {'child_jobs':jobids}
def generate_images_parallel(outdir,id_list):
jobids = []
for (i,id) in enumerate(id_list):
url = 'http://50.19.109.25:9999/3dmodels?query={"id":"' + id + '"}'
y = json.loads(urllib.urlopen(url).read())[0]
x = {'bg_id':'INTERIOR_10SN.tdl',
'model_params':[{'model_id':y['id'],
'rxy':y.get('canonical_view',{}).get('rxy',0),
'rxz':y.get('canonical_view',{}).get('rxz',0),
'ryz':y.get('canonical_view',{}).get('ryz',0)
}]
}
jobid = qsub(generate_single_image,(x,outdir),opstring='-pe orte 2 -l qname=rendering.q -o /home/render -e /home/render')
jobids.append(jobid)
return {'child_jobs':jobids}
def generate_images_parallel(outfile,im_hash,config_gen):
conn = pm.Connection(document_class = SON)
db = conn[DB_NAME]
im_coll = db['images.files']
im_fs = gridfs.GridFS(db,'images')
remove_existing(im_coll,im_fs,im_hash)
X = rendering.config_gen(config_gen)
jobids = []
for (i,x) in enumerate(X):
x['image']['generator'] = config_gen['images']['generator']
jobid = qsub(generate_and_insert_single_image,(x,im_hash),opstring='-pe orte 2 -l qname=rendering.q -o /home/render -e /home/render')
jobids.append(jobid)
createCertificateDict(outfile,{'image_hash':im_hash,'args':config_gen})
return {'child_jobs':jobids}
def extract_and_evaluate_parallel(outfile,image_certificate_file,model_certificate_file,cpath,convolve_func_name,task,ext_hash):
(model_configs, image_config_gen, model_hash, image_hash, task_list,
perf_col, split_coll, split_fs, splitperf_coll, splitperf_fs) = prepare_extract_and_evaluate(ext_hash,
image_certificate_file,
model_certificate_file,
task)
jobids = []
if convolve_func_name == 'numpy':
opstring = '-l qname=extraction_cpu.q'
elif convolve_func_name == 'pyfft':
opstring = '-l qname=extraction_gpu.q -o /home/render -e /home/render'
for m in model_configs:
print('Evaluating model',m)
for task in task_list:
classifier_kwargs = task.get('classifier_kwargs',{})
print('task',task)
splits = generate_splits(task,image_hash,'images')
for (ind,split) in enumerate(splits):
put_in_split(split,image_config_gen,m,task,ext_hash,ind,split_fs)
jobid = qsub(extract_and_evaluate_parallel_core,(image_config_gen,m,task,ext_hash,ind,convolve_func_name),opstring=opstring)
jobids.append(jobid)
print(jobids)
statuses = wait_and_get_statuses(jobids)
for m in model_configs:
print('Evaluating model',m)
for task in task_list:
split_results = get_most_recent_files(splitperf_coll,{'__hash__':ext_hash,'task':son_escape(task),'model':m['config']['model'],'images':son_escape(image_config_gen['images'])})
put_in_performance(split_results,image_config_gen,m,model_hash,image_hash,perf_col,task,ext_hash)
createCertificateDict(outfile,{'image_file':image_certificate_file,'models_file':model_certificate_file})
