Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
G
GenVQA
Manage
Activity
Members
Code
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Deploy
Releases
Model registry
Analyze
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Vision
GenVQA
Graph
f07ae5422dc2cb5345be44c398064dd053817c3b
Select Git revision
Branches
20
GenVQA_obj_atr_ans_mil
acc_vs_qa_rarity
acc_vs_ratio
add_test_eval
avg_reg_feats
backup_simple_avg_reg_feat
bn_before_feat_add
cached_features
classifiers
discon_wordvec_classifiers
elementwise_multiply
extract_word_vectors
fix_genome_split
fix_label_synonyms
gpu_machine_setup
inner_product_selection
large_batch_size
large_cap_obj_atr
localization_layer
make_fast
20 results
You can move around the graph by using the arrow keys.
Begin with the selected commit
Created with Raphaël 2.2.0
11
Mar
11
Nov
9
8
7
6
4
3
2
1
29
Oct
28
27
24
23
21
19
18
17
14
13
12
11
10
9
8
6
4
30
Sep
28
27
25
24
23
21
19
17
16
14
13
12
9
4
2
1
29
Aug
26
25
24
20
19
15
11
8
5
4
3
1
26
Jul
24
11
6
24
Jun
21
14
9
8
6
4
19
May
15
9
2
28
Apr
25
21
17
12
10
7
4
30
Mar
29
28
25
23
21
20
18
17
16
9
7
5
4
21
Feb
18
merging scripts and modifications for train + val training
resnet_50_train…
resnet_50_train_val
constants_crunchy update
acc_vs_qa_rarity
acc_vs_qa_rarity
acc vs rarity final version
acc_vs_rarity
change accuracy to top-20 instead of top-100
acc_vs_ratio
acc_vs_ratio
compute average mentions in freq_acc_plot_obj
play with different binning schemes in ratio_acc_plot_obj
ratio_acc_plot_obj
count_objects_in_vqa
constants_crunchy update
constants_vision_gpu_1 update
cmp_obj_freq_acc
constants_crunchy update
rel_bin_feats
rel_bin_feats
ratio_acc_plot_obj
count_objects_in_vqa
extract word vectors
extract_word_ve…
extract_word_vectors
add new constants_crunchy
no_single_feats…
no_single_feats_pretrained
make sure word vecs and obj graph are being backproped into
noun adj identity and yes_no_num in ans eval
no_single_feats…
no_single_feats_discon
lower learning rate for obj/atr graph and word vectors
comment of printing of max_ans_noun_score in ans train
comment of printing of max_ans_noun_score in ans train
use fc instead of word vectors for obj atr classification
eval and visualize script for the interpreted evaluation
Merge branch 'eval_interpret' into rel_bin_feats
fix , bug
eval_interpret
make visualization darker
answer eval
eval_interpret
single_features
single_features
make visualization darker
answer eval
answer eval
fix bug in yes no feat
updated constants_vision_gpu_1
add all rel and yes no num features
constants_vision_gpu_1 update
no_single_feats
no_single_feats
zero out rel feat, yes no features already zeroed out
ans select best model
changed k to 100 in obj atr eval
Loading