def testSubsetFiles(meta_val_file,meta_synset_words_file,sub_val_file,sub_synset_words_file,excluded_ids_all_file,removePath):
meta_list_files_val,meta_idx_val=zip(*imagenet.readLabelsFile(meta_val_file));
meta_val_idx_mapped=[int(idx_curr) for idx_curr in meta_idx_val];
meta_val_idx_mapped=np.array(meta_val_idx_mapped);
t=time.time();
meta_ids,meta_labels=zip(*imagenet.readLabelsFile(meta_synset_words_file));
meta_ids=np.array(meta_ids);
meta_val_ids_mapped=np.empty((len(meta_list_files_val),),dtype='object');
for idx_curr in np.unique(meta_val_idx_mapped):
idx_rel=meta_val_idx_mapped==idx_curr;
meta_val_ids_mapped[idx_rel]=meta_ids[idx_curr];
meta_val_ids_mapped=list(meta_val_ids_mapped);
print time.time()-t
sub_list_files_val,sub_idx_val=zip(*imagenet.readLabelsFile(sub_val_file));
sub_val_idx_mapped=[int(idx_curr) for idx_curr in sub_idx_val];
excluded_ids_all=imagenet.readSynsetsFile(excluded_ids_all_file);
#all excluded files are excluded
check_exclude=testFilesExcludedProperly(sub_list_files_val,meta_list_files_val,meta_val_ids_mapped,excluded_ids_all)
#all included files are included
check_include=testFilesIncludedProperly(sub_list_files_val,meta_list_files_val,meta_val_ids_mapped,excluded_ids_all)
#all indices labeling is consistent
check_labeling=testFilesLabeledProperly(sub_val_idx_mapped,sub_list_files_val,meta_val_idx_mapped,meta_list_files_val)
return check_exclude,check_include,check_labeling;
def testSubsetFiles(meta_val_file, meta_synset_words_file, sub_val_file,
sub_synset_words_file, excluded_ids_all_file, removePath):
meta_list_files_val, meta_idx_val = zip(
*imagenet.readLabelsFile(meta_val_file))
meta_val_idx_mapped = [int(idx_curr) for idx_curr in meta_idx_val]
meta_val_idx_mapped = np.array(meta_val_idx_mapped)
t = time.time()
meta_ids, meta_labels = zip(
*imagenet.readLabelsFile(meta_synset_words_file))
meta_ids = np.array(meta_ids)
meta_val_ids_mapped = np.empty((len(meta_list_files_val), ),
dtype='object')
for idx_curr in np.unique(meta_val_idx_mapped):
idx_rel = meta_val_idx_mapped == idx_curr
meta_val_ids_mapped[idx_rel] = meta_ids[idx_curr]
meta_val_ids_mapped = list(meta_val_ids_mapped)
print time.time() - t
sub_list_files_val, sub_idx_val = zip(
*imagenet.readLabelsFile(sub_val_file))
sub_val_idx_mapped = [int(idx_curr) for idx_curr in sub_idx_val]
excluded_ids_all = imagenet.readSynsetsFile(excluded_ids_all_file)
#all excluded files are excluded
check_exclude = testFilesExcludedProperly(sub_list_files_val,
meta_list_files_val,
meta_val_ids_mapped,
excluded_ids_all)
#all included files are included
check_include = testFilesIncludedProperly(sub_list_files_val,
meta_list_files_val,
meta_val_ids_mapped,
excluded_ids_all)
#all indices labeling is consistent
check_labeling = testFilesLabeledProperly(sub_val_idx_mapped,
sub_list_files_val,
meta_val_idx_mapped,
meta_list_files_val)
return check_exclude, check_include, check_labeling
def main():
meta_dir = "../../data/ilsvrc12/"
meta_synsets_file = os.path.join(meta_dir, "synsets.txt")
meta_synset_words_file = os.path.join(meta_dir, "synset_words.txt")
meta_train_file = os.path.join(meta_dir, "train.txt")
meta_val_file = os.path.join(meta_dir, "val.txt")
sub_dir = "/disk2/novemberExperiments/network_no_pascal"
sub_synsets_file = os.path.join(sub_dir, "synsets.txt")
sub_synset_words_file = os.path.join(sub_dir, "synset_words.txt")
sub_train_file = os.path.join(sub_dir, "train.txt")
sub_val_file = os.path.join(sub_dir, "val.txt")
excluded_ids_all_file = os.path.join(sub_dir, "to_exclude_all.txt")
# checks=tests.testSubsetFiles(meta_val_file,meta_synset_words_file,sub_val_file,sub_synset_words_file,excluded_ids_all_file,True);
# print checks
checks = tests.testSubsetFiles(
meta_train_file,
meta_synset_words_file,
sub_train_file,
sub_synset_words_file,
excluded_ids_all_file,
removePath=False,
)
print checks
return
in_dir = "/disk2/octoberExperiments/nn_performance_without_pascal"
out_dir = "/disk2/novemberExperiments/network_no_pascal"
parent_synset_words_file = os.path.join(in_dir, "synset_words.txt")
parent_val_file = os.path.join(in_dir, "new_val.txt")
parent_train_file = os.path.join(in_dir, "train.txt")
parent_synset_file = os.path.join(in_dir, "synsets.txt")
new_val_file = os.path.join(out_dir, "val.txt")
new_train_file = os.path.join(out_dir, "train.txt")
new_synset_file = os.path.join(out_dir, "synsets.txt")
new_synset_words_file = os.path.join(out_dir, "synset_words.txt")
to_exclude_text_file = os.path.join(out_dir, "to_exclude.txt")
to_exclude_meta_file = os.path.join(out_dir, "to_exclude_all.txt")
to_exclude_2 = imagenet.readSynsetsFile(to_exclude_text_file)
to_exclude_1 = imagenet.readSynsetsFile(os.path.join(in_dir, "to_exclude.txt"))
print len(to_exclude_1), len(to_exclude_2)
to_exclude_all = to_exclude_1 + to_exclude_2
with open(to_exclude_meta_file, "wb") as f:
for id_curr in to_exclude_all:
f.write(id_curr + "\n")
print to_exclude_meta_file
return
params = createParams("setUpFilesForTrainingOnSubset")
params = params(
parent_synset_words_file=parent_synset_words_file,
parent_val_file=parent_val_file,
parent_train_file=parent_train_file,
parent_synset_file=parent_synset_file,
new_val_file=new_val_file,
new_train_file=new_train_file,
new_synset_file=new_synset_file,
new_synset_words_file=new_synset_words_file,
to_exclude_text_file=to_exclude_text_file,
)
script_setUpFilesForTrainingOnSubset(params)
