Hugging Face's logo

ooliverz
/
git-large-r-coco-IDB_ADv1_COCOv4

Image-Text-to-Text
Transformers
TensorBoard
Safetensors
git
Generated from Trainer
Model card Files Metrics Community
ooliverz commited on
Commit
d4551a5
·
verified ·
1 Parent(s): 5ab05aa

End of training

Browse files
Files changed (2) hide show
  1. README.md +44 -45
  2. model.safetensors +1 -1
README.md CHANGED
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [microsoft/git-large-r-coco](https://huggingface.co/microsoft/git-large-r-coco) on the imagefolder dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 2.3179
22
- - Meteor Score: {'meteor': 0.5631808848167792}
23
 
24
  ## Model description
25
 
@@ -46,54 +46,53 @@ The following hyperparameters were used during training:
46
  - total_train_batch_size: 256
47
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: cosine
49
- - lr_scheduler_warmup_steps: 500
50
  - num_epochs: 200
51
  - mixed_precision_training: Native AMP
52
 
53
  ### Training results
54
 
55
- | Training Loss | Epoch | Step | Validation Loss | Meteor Score |
56
- |:-------------:|:-----:|:----:|:---------------:|:------------------------------:|
57
- | 4.9891 | 5.0 | 5 | 4.8898 | {'meteor': 0.4977448904730416} |
58
- | 4.9871 | 10.0 | 10 | 4.8871 | {'meteor': 0.4986783896823098} |
59
- | 4.9807 | 15.0 | 15 | 4.8787 | {'meteor': 0.5039923089582024} |
60
- | 4.9703 | 20.0 | 20 | 4.8695 | {'meteor': 0.5007229587447326} |
61
- | 4.9552 | 25.0 | 25 | 4.8554 | {'meteor': 0.5040915299837979} |
62
- | 4.937 | 30.0 | 30 | 4.8380 | {'meteor': 0.5048006043123706} |
63
- | 4.9144 | 35.0 | 35 | 4.8172 | {'meteor': 0.5053354754203138} |
64
- | 4.8884 | 40.0 | 40 | 4.7941 | {'meteor': 0.5052519582427984} |
65
- | 4.8595 | 45.0 | 45 | 4.7648 | {'meteor': 0.5155678605148059} |
66
- | 4.8273 | 50.0 | 50 | 4.7351 | {'meteor': 0.5141758460997891} |
67
- | 4.7906 | 55.0 | 55 | 4.7006 | {'meteor': 0.5161210615484346} |
68
- | 4.7514 | 60.0 | 60 | 4.6661 | {'meteor': 0.5227861070655556} |
69
- | 4.7102 | 65.0 | 65 | 4.6277 | {'meteor': 0.5254243476755014} |
70
- | 4.6657 | 70.0 | 70 | 4.5848 | {'meteor': 0.5268576504702123} |
71
- | 4.6174 | 75.0 | 75 | 4.5400 | {'meteor': 0.5327203741360704} |
72
- | 4.5668 | 80.0 | 80 | 4.4927 | {'meteor': 0.5328385720589747} |
73
- | 4.5133 | 85.0 | 85 | 4.4344 | {'meteor': 0.5379080722700745} |
74
- | 4.4563 | 90.0 | 90 | 4.3847 | {'meteor': 0.5351712288811252} |
75
- | 4.3961 | 95.0 | 95 | 4.3223 | {'meteor': 0.5459124584203718} |
76
- | 4.3328 | 100.0 | 100 | 4.2619 | {'meteor': 0.5450854025913955} |
77
- | 4.2664 | 105.0 | 105 | 4.1984 | {'meteor': 0.5396980345283648} |
78
- | 4.1973 | 110.0 | 110 | 4.1254 | {'meteor': 0.5436701507478281} |
79
- | 4.1255 | 115.0 | 115 | 4.0597 | {'meteor': 0.5443380961541314} |
80
- | 4.0497 | 120.0 | 120 | 3.9792 | {'meteor': 0.5501741486578466} |
81
- | 3.9694 | 125.0 | 125 | 3.9008 | {'meteor': 0.549708872514102} |
82
- | 3.8871 | 130.0 | 130 | 3.8238 | {'meteor': 0.548104982356801} |
83
- | 3.8014 | 135.0 | 135 | 3.7364 | {'meteor': 0.5554349932565801} |
84
- | 3.7128 | 140.0 | 140 | 3.6483 | {'meteor': 0.55742212703008} |
85
- | 3.6206 | 145.0 | 145 | 3.5535 | {'meteor': 0.554548892112528} |
86
- | 3.5247 | 150.0 | 150 | 3.4614 | {'meteor': 0.5577247301342534} |
87
- | 3.4264 | 155.0 | 155 | 3.3604 | {'meteor': 0.5567178684938024} |
88
- | 3.3247 | 160.0 | 160 | 3.2577 | {'meteor': 0.5542593803491448} |
89
- | 3.2182 | 165.0 | 165 | 3.1494 | {'meteor': 0.5517431992593748} |
90
- | 3.1093 | 170.0 | 170 | 3.0413 | {'meteor': 0.5609162269594461} |
91
- | 2.997 | 175.0 | 175 | 2.9274 | {'meteor': 0.5595566101373656} |
92
- | 2.8804 | 180.0 | 180 | 2.8130 | {'meteor': 0.5594850153568144} |
93
- | 2.7608 | 185.0 | 185 | 2.6924 | {'meteor': 0.5605989110400809} |
94
- | 2.6389 | 190.0 | 190 | 2.5684 | {'meteor': 0.5608506949746257} |
95
- | 2.5144 | 195.0 | 195 | 2.4447 | {'meteor': 0.5655230759592001} |
96
- | 2.3862 | 200.0 | 200 | 2.3179 | {'meteor': 0.5631808848167792} |
97
 
98
 
99
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [microsoft/git-large-r-coco](https://huggingface.co/microsoft/git-large-r-coco) on the imagefolder dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.0985
22
+ - Meteor Score: {'meteor': 0.6045977816224402}
23
 
24
  ## Model description
25
 
 
46
  - total_train_batch_size: 256
47
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: cosine
 
49
  - num_epochs: 200
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Meteor Score |
55
+ |:-------------:|:-----:|:----:|:---------------:|:-------------------------------:|
56
+ | 3.0153 | 5.0 | 5 | 2.9012 | {'meteor': 0.11360374562500358} |
57
+ | 2.6775 | 10.0 | 10 | 1.9996 | {'meteor': 0.3522292411621663} |
58
+ | 1.9479 | 15.0 | 15 | 1.6433 | {'meteor': 0.4724950837310155} |
59
+ | 1.6034 | 20.0 | 20 | 1.3467 | {'meteor': 0.5315599885135005} |
60
+ | 1.3119 | 25.0 | 25 | 1.0907 | {'meteor': 0.5441775609077182} |
61
+ | 1.0551 | 30.0 | 30 | 0.8677 | {'meteor': 0.5627206125493585} |
62
+ | 0.8324 | 35.0 | 35 | 0.6841 | {'meteor': 0.5644140822761644} |
63
+ | 0.6466 | 40.0 | 40 | 0.5342 | {'meteor': 0.5714878241998325} |
64
+ | 0.4953 | 45.0 | 45 | 0.4171 | {'meteor': 0.5723224079693889} |
65
+ | 0.3758 | 50.0 | 50 | 0.3280 | {'meteor': 0.575759183770385} |
66
+ | 0.2832 | 55.0 | 55 | 0.2617 | {'meteor': 0.582651143305794} |
67
+ | 0.2147 | 60.0 | 60 | 0.2141 | {'meteor': 0.5823617638426744} |
68
+ | 0.1637 | 65.0 | 65 | 0.1841 | {'meteor': 0.587684204036701} |
69
+ | 0.1283 | 70.0 | 70 | 0.1510 | {'meteor': 0.5995580706928448} |
70
+ | 0.097 | 75.0 | 75 | 0.1344 | {'meteor': 0.5935823756573874} |
71
+ | 0.0745 | 80.0 | 80 | 0.1212 | {'meteor': 0.5971126553853467} |
72
+ | 0.0576 | 85.0 | 85 | 0.1175 | {'meteor': 0.6063465199710643} |
73
+ | 0.0482 | 90.0 | 90 | 0.1095 | {'meteor': 0.6256289206454945} |
74
+ | 0.041 | 95.0 | 95 | 0.1021 | {'meteor': 0.611542182926066} |
75
+ | 0.0332 | 100.0 | 100 | 0.0967 | {'meteor': 0.6332339587712169} |
76
+ | 0.0263 | 105.0 | 105 | 0.0974 | {'meteor': 0.6217214977486667} |
77
+ | 0.021 | 110.0 | 110 | 0.0969 | {'meteor': 0.6205827762386601} |
78
+ | 0.0175 | 115.0 | 115 | 0.0974 | {'meteor': 0.6149467166555415} |
79
+ | 0.015 | 120.0 | 120 | 0.0969 | {'meteor': 0.6141265010864785} |
80
+ | 0.0132 | 125.0 | 125 | 0.0968 | {'meteor': 0.6148825585221068} |
81
+ | 0.0119 | 130.0 | 130 | 0.0947 | {'meteor': 0.6099154157322018} |
82
+ | 0.0109 | 135.0 | 135 | 0.0962 | {'meteor': 0.6091804819899924} |
83
+ | 0.0101 | 140.0 | 140 | 0.0967 | {'meteor': 0.6079888261097917} |
84
+ | 0.0096 | 145.0 | 145 | 0.0983 | {'meteor': 0.6084492863562021} |
85
+ | 0.0091 | 150.0 | 150 | 0.0982 | {'meteor': 0.6061886051922543} |
86
+ | 0.0088 | 155.0 | 155 | 0.0985 | {'meteor': 0.6057257284449861} |
87
+ | 0.0085 | 160.0 | 160 | 0.0988 | {'meteor': 0.603016234486876} |
88
+ | 0.0083 | 165.0 | 165 | 0.0984 | {'meteor': 0.6042894593740327} |
89
+ | 0.0082 | 170.0 | 170 | 0.0984 | {'meteor': 0.6043600622898042} |
90
+ | 0.0081 | 175.0 | 175 | 0.0985 | {'meteor': 0.6043742843963279} |
91
+ | 0.008 | 180.0 | 180 | 0.0985 | {'meteor': 0.6041252208578323} |
92
+ | 0.008 | 185.0 | 185 | 0.0986 | {'meteor': 0.6043322953582039} |
93
+ | 0.0079 | 190.0 | 190 | 0.0986 | {'meteor': 0.6047551630278664} |
94
+ | 0.0079 | 195.0 | 195 | 0.0985 | {'meteor': 0.6048157030511828} |
95
+ | 0.0079 | 200.0 | 200 | 0.0985 | {'meteor': 0.6045977816224402} |
96
 
97
 
98
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:18bcac3599311af608500c46b9610a5b17915fb47eafff8059068299cbb09726
3
  size 1576851440
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dff81bc5cb71924ce28776519f89f5567e38f7f18b698a2a59667eec5b40048e
3
  size 1576851440