Hugging Face's logo

jena-shreyas
/
florence_ft

Image-Text-to-Text
Transformers
TensorBoard
Safetensors
florence2
Generated from Trainer
custom_code
Model card Files
xet
 Metrics Community
jena-shreyas commited on
Commit
78308fd
·
verified ·
1 Parent(s): cd2ebc0

Model save

Browse files
Files changed (3) hide show
  1. README.md +94 -115
  2. generation_config.json +1 -1
  3. model.safetensors +1 -1
README.md CHANGED
@@ -1,9 +1,8 @@
1
  ---
 
2
  base_model: HuggingFaceM4/Florence-2-DocVQA
3
  tags:
4
  - generated_from_trainer
5
- metrics:
6
- - accuracy
7
  model-index:
8
  - name: florence_ft
9
  results: []
@@ -12,13 +11,11 @@ model-index:
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
  should probably proofread and complete it, then remove this comment. -->
14
 
15
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/jenashreyas/florence_entity_extraction_ft/runs/zobmh6l8)
16
  # florence_ft
17
 
18
  This model is a fine-tuned version of [HuggingFaceM4/Florence-2-DocVQA](https://huggingface.co/HuggingFaceM4/Florence-2-DocVQA) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 1.0500
21
- - Accuracy: 0.0
22
 
23
  ## Model description
24
 
@@ -37,124 +34,106 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - learning_rate: 2e-05
41
- - train_batch_size: 32
42
- - eval_batch_size: 32
43
  - seed: 42
 
 
44
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
  - lr_scheduler_type: linear
46
- - lr_scheduler_warmup_steps: 2
47
- - num_epochs: 100
48
 
49
  ### Training results
50
 
51
- | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
- |:-------------:|:-----:|:----:|:---------------:|:--------:|
53
- | No log | 1.0 | 7 | 1.4280 | 0.0 |
54
- | 2.3602 | 2.0 | 14 | 1.2109 | 0.0 |
55
- | 1.0245 | 3.0 | 21 | 1.2202 | 0.0 |
56
- | 1.0245 | 4.0 | 28 | 1.2606 | 0.0 |
57
- | 0.929 | 5.0 | 35 | 1.2130 | 0.0 |
58
- | 0.8584 | 6.0 | 42 | 1.1525 | 0.0 |
59
- | 0.8584 | 7.0 | 49 | 1.0689 | 0.0 |
60
- | 0.7889 | 8.0 | 56 | 1.0324 | 0.0 |
61
- | 0.7719 | 9.0 | 63 | 1.0274 | 0.0 |
62
- | 0.7284 | 10.0 | 70 | 1.0232 | 0.0 |
63
- | 0.7284 | 11.0 | 77 | 1.0275 | 0.0 |
64
- | 0.6993 | 12.0 | 84 | 1.0357 | 0.0 |
65
- | 0.688 | 13.0 | 91 | 1.0444 | 0.0 |
66
- | 0.688 | 14.0 | 98 | 1.0427 | 0.0 |
67
- | 0.6764 | 15.0 | 105 | 1.0408 | 0.0 |
68
- | 0.6521 | 16.0 | 112 | 1.0217 | 0.0 |
69
- | 0.6521 | 17.0 | 119 | 1.0021 | 0.0 |
70
- | 0.6511 | 18.0 | 126 | 1.0037 | 0.0 |
71
- | 0.637 | 19.0 | 133 | 1.0241 | 0.0 |
72
- | 0.6348 | 20.0 | 140 | 1.0314 | 0.0 |
73
- | 0.6348 | 21.0 | 147 | 1.0510 | 0.0 |
74
- | 0.6165 | 22.0 | 154 | 1.0596 | 0.0 |
75
- | 0.6245 | 23.0 | 161 | 1.0526 | 0.0 |
76
- | 0.6245 | 24.0 | 168 | 1.0489 | 0.0 |
77
- | 0.6107 | 25.0 | 175 | 1.0402 | 0.0 |
78
- | 0.6012 | 26.0 | 182 | 1.0453 | 0.0 |
79
- | 0.6012 | 27.0 | 189 | 1.0450 | 0.0 |
80
- | 0.5995 | 28.0 | 196 | 1.0416 | 0.0 |
81
- | 0.5975 | 29.0 | 203 | 1.0469 | 0.0 |
82
- | 0.5834 | 30.0 | 210 | 1.0590 | 0.0 |
83
- | 0.5834 | 31.0 | 217 | 1.0518 | 0.0 |
84
- | 0.585 | 32.0 | 224 | 1.0644 | 0.0 |
85
- | 0.5846 | 33.0 | 231 | 1.0692 | 0.0 |
86
- | 0.5846 | 34.0 | 238 | 1.0526 | 0.0 |
87
- | 0.5842 | 35.0 | 245 | 1.0608 | 0.0 |
88
- | 0.5783 | 36.0 | 252 | 1.0644 | 0.0 |
89
- | 0.5783 | 37.0 | 259 | 1.0479 | 0.0 |
90
- | 0.5899 | 38.0 | 266 | 1.0503 | 0.0 |
91
- | 0.5766 | 39.0 | 273 | 1.0502 | 0.0 |
92
- | 0.575 | 40.0 | 280 | 1.0606 | 0.0 |
93
- | 0.575 | 41.0 | 287 | 1.0568 | 0.0 |
94
- | 0.569 | 42.0 | 294 | 1.0587 | 0.0 |
95
- | 0.5673 | 43.0 | 301 | 1.0670 | 0.0 |
96
- | 0.5673 | 44.0 | 308 | 1.0699 | 0.0 |
97
- | 0.5663 | 45.0 | 315 | 1.0731 | 0.0 |
98
- | 0.5681 | 46.0 | 322 | 1.0819 | 0.0 |
99
- | 0.5681 | 47.0 | 329 | 1.0885 | 0.0 |
100
- | 0.5578 | 48.0 | 336 | 1.0928 | 0.0 |
101
- | 0.5641 | 49.0 | 343 | 1.0937 | 0.0 |
102
- | 0.5657 | 50.0 | 350 | 1.0815 | 0.0 |
103
- | 0.5657 | 51.0 | 357 | 1.0746 | 0.0 |
104
- | 0.5583 | 52.0 | 364 | 1.0672 | 0.0 |
105
- | 0.5664 | 53.0 | 371 | 1.0643 | 0.0 |
106
- | 0.5664 | 54.0 | 378 | 1.0648 | 0.0 |
107
- | 0.5614 | 55.0 | 385 | 1.0605 | 0.0 |
108
- | 0.5592 | 56.0 | 392 | 1.0610 | 0.0 |
109
- | 0.5592 | 57.0 | 399 | 1.0587 | 0.0 |
110
- | 0.5542 | 58.0 | 406 | 1.0614 | 0.0 |
111
- | 0.5629 | 59.0 | 413 | 1.0573 | 0.0 |
112
- | 0.549 | 60.0 | 420 | 1.0573 | 0.0 |
113
- | 0.549 | 61.0 | 427 | 1.0559 | 0.0 |
114
- | 0.5573 | 62.0 | 434 | 1.0581 | 0.0 |
115
- | 0.5656 | 63.0 | 441 | 1.0548 | 0.0 |
116
- | 0.5656 | 64.0 | 448 | 1.0515 | 0.0 |
117
- | 0.5489 | 65.0 | 455 | 1.0517 | 0.0 |
118
- | 0.5531 | 66.0 | 462 | 1.0514 | 0.0 |
119
- | 0.5531 | 67.0 | 469 | 1.0546 | 0.0 |
120
- | 0.5463 | 68.0 | 476 | 1.0553 | 0.0 |
121
- | 0.5527 | 69.0 | 483 | 1.0580 | 0.0 |
122
- | 0.554 | 70.0 | 490 | 1.0559 | 0.0 |
123
- | 0.554 | 71.0 | 497 | 1.0555 | 0.0 |
124
- | 0.5524 | 72.0 | 504 | 1.0566 | 0.0 |
125
- | 0.5498 | 73.0 | 511 | 1.0560 | 0.0 |
126
- | 0.5498 | 74.0 | 518 | 1.0569 | 0.0 |
127
- | 0.5592 | 75.0 | 525 | 1.0565 | 0.0 |
128
- | 0.561 | 76.0 | 532 | 1.0515 | 0.0 |
129
- | 0.561 | 77.0 | 539 | 1.0494 | 0.0 |
130
- | 0.5473 | 78.0 | 546 | 1.0507 | 0.0 |
131
- | 0.5493 | 79.0 | 553 | 1.0506 | 0.0 |
132
- | 0.5532 | 80.0 | 560 | 1.0491 | 0.0 |
133
- | 0.5532 | 81.0 | 567 | 1.0498 | 0.0 |
134
- | 0.5484 | 82.0 | 574 | 1.0481 | 0.0 |
135
- | 0.5523 | 83.0 | 581 | 1.0511 | 0.0 |
136
- | 0.5523 | 84.0 | 588 | 1.0498 | 0.0 |
137
- | 0.5496 | 85.0 | 595 | 1.0504 | 0.0 |
138
- | 0.5485 | 86.0 | 602 | 1.0499 | 0.0 |
139
- | 0.5485 | 87.0 | 609 | 1.0501 | 0.0 |
140
- | 0.5418 | 88.0 | 616 | 1.0501 | 0.0 |
141
- | 0.5547 | 89.0 | 623 | 1.0521 | 0.0 |
142
- | 0.5435 | 90.0 | 630 | 1.0511 | 0.0 |
143
- | 0.5435 | 91.0 | 637 | 1.0502 | 0.0 |
144
- | 0.5488 | 92.0 | 644 | 1.0506 | 0.0 |
145
- | 0.5472 | 93.0 | 651 | 1.0506 | 0.0 |
146
- | 0.5472 | 94.0 | 658 | 1.0503 | 0.0 |
147
- | 0.5521 | 95.0 | 665 | 1.0507 | 0.0 |
148
- | 0.5485 | 96.0 | 672 | 1.0509 | 0.0 |
149
- | 0.5485 | 97.0 | 679 | 1.0500 | 0.0 |
150
- | 0.5611 | 98.0 | 686 | 1.0514 | 0.0 |
151
- | 0.5517 | 99.0 | 693 | 1.0508 | 0.0 |
152
- | 0.5574 | 100.0 | 700 | 1.0500 | 0.0 |
153
 
154
 
155
  ### Framework versions
156
 
157
- - Transformers 4.42.0
158
- - Pytorch 2.1.2+cu121
159
- - Datasets 2.16.1
160
  - Tokenizers 0.19.1
 
1
  ---
2
+ library_name: transformers
3
  base_model: HuggingFaceM4/Florence-2-DocVQA
4
  tags:
5
  - generated_from_trainer
 
 
6
  model-index:
7
  - name: florence_ft
8
  results: []
 
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
  should probably proofread and complete it, then remove this comment. -->
13
 
 
14
  # florence_ft
15
 
16
  This model is a fine-tuned version of [HuggingFaceM4/Florence-2-DocVQA](https://huggingface.co/HuggingFaceM4/Florence-2-DocVQA) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 1.0833
 
19
 
20
  ## Model description
21
 
 
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
+ - learning_rate: 5e-06
38
+ - train_batch_size: 8
39
+ - eval_batch_size: 8
40
  - seed: 42
41
+ - gradient_accumulation_steps: 4
42
+ - total_train_batch_size: 32
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
+ - lr_scheduler_warmup_ratio: 0.05
46
+ - num_epochs: 1
47
 
48
  ### Training results
49
 
50
+ | Training Loss | Epoch | Step | Validation Loss |
51
+ |:-------------:|:------:|:----:|:---------------:|
52
+ | 4.4629 | 0.0123 | 25 | 4.6140 |
53
+ | 4.0165 | 0.0245 | 50 | 3.9075 |
54
+ | 3.0887 | 0.0368 | 75 | 2.4186 |
55
+ | 1.3752 | 0.0491 | 100 | 1.4240 |
56
+ | 1.1205 | 0.0613 | 125 | 1.2705 |
57
+ | 1.0809 | 0.0736 | 150 | 1.2144 |
58
+ | 1.0946 | 0.0859 | 175 | 1.1813 |
59
+ | 1.0311 | 0.0982 | 200 | 1.1653 |
60
+ | 1.0611 | 0.1104 | 225 | 1.1503 |
61
+ | 1.0209 | 0.1227 | 250 | 1.1423 |
62
+ | 1.052 | 0.1350 | 275 | 1.1384 |
63
+ | 1.0129 | 0.1472 | 300 | 1.1273 |
64
+ | 0.9764 | 0.1595 | 325 | 1.1218 |
65
+ | 0.9707 | 0.1718 | 350 | 1.1155 |
66
+ | 1.0024 | 0.1840 | 375 | 1.1151 |
67
+ | 1.0446 | 0.1963 | 400 | 1.1112 |
68
+ | 0.9691 | 0.2086 | 425 | 1.1081 |
69
+ | 1.0018 | 0.2209 | 450 | 1.1040 |
70
+ | 0.9806 | 0.2331 | 475 | 1.0989 |
71
+ | 1.0555 | 0.2454 | 500 | 1.0980 |
72
+ | 0.9778 | 0.2577 | 525 | 1.0981 |
73
+ | 0.988 | 0.2699 | 550 | 1.0962 |
74
+ | 0.988 | 0.2822 | 575 | 1.0939 |
75
+ | 0.9572 | 0.2945 | 600 | 1.0969 |
76
+ | 0.9802 | 0.3067 | 625 | 1.0952 |
77
+ | 0.9504 | 0.3190 | 650 | 1.0933 |
78
+ | 1.0194 | 0.3313 | 675 | 1.0948 |
79
+ | 0.9697 | 0.3436 | 700 | 1.0935 |
80
+ | 0.96 | 0.3558 | 725 | 1.0903 |
81
+ | 0.9665 | 0.3681 | 750 | 1.0924 |
82
+ | 0.9895 | 0.3804 | 775 | 1.0920 |
83
+ | 1.004 | 0.3926 | 800 | 1.0914 |
84
+ | 1.0054 | 0.4049 | 825 | 1.0909 |
85
+ | 0.9514 | 0.4172 | 850 | 1.0890 |
86
+ | 0.9996 | 0.4294 | 875 | 1.0906 |
87
+ | 0.99 | 0.4417 | 900 | 1.0896 |
88
+ | 0.9427 | 0.4540 | 925 | 1.0887 |
89
+ | 1.0014 | 0.4663 | 950 | 1.0883 |
90
+ | 0.9639 | 0.4785 | 975 | 1.0864 |
91
+ | 1.0073 | 0.4908 | 1000 | 1.0877 |
92
+ | 0.9895 | 0.5031 | 1025 | 1.0863 |
93
+ | 0.9594 | 0.5153 | 1050 | 1.0841 |
94
+ | 0.9559 | 0.5276 | 1075 | 1.0849 |
95
+ | 1.0034 | 0.5399 | 1100 | 1.0849 |
96
+ | 0.9795 | 0.5521 | 1125 | 1.0844 |
97
+ | 0.9661 | 0.5644 | 1150 | 1.0834 |
98
+ | 0.9533 | 0.5767 | 1175 | 1.0830 |
99
+ | 0.976 | 0.5890 | 1200 | 1.0830 |
100
+ | 0.9932 | 0.6012 | 1225 | 1.0846 |
101
+ | 1.0067 | 0.6135 | 1250 | 1.0861 |
102
+ | 0.9543 | 0.6258 | 1275 | 1.0854 |
103
+ | 0.9733 | 0.6380 | 1300 | 1.0844 |
104
+ | 0.9673 | 0.6503 | 1325 | 1.0837 |
105
+ | 0.9378 | 0.6626 | 1350 | 1.0837 |
106
+ | 0.9713 | 0.6748 | 1375 | 1.0840 |
107
+ | 0.9913 | 0.6871 | 1400 | 1.0838 |
108
+ | 0.9302 | 0.6994 | 1425 | 1.0837 |
109
+ | 0.9873 | 0.7117 | 1450 | 1.0836 |
110
+ | 0.9618 | 0.7239 | 1475 | 1.0835 |
111
+ | 1.0042 | 0.7362 | 1500 | 1.0835 |
112
+ | 0.9627 | 0.7485 | 1525 | 1.0827 |
113
+ | 0.9635 | 0.7607 | 1550 | 1.0827 |
114
+ | 0.9658 | 0.7730 | 1575 | 1.0828 |
115
+ | 0.9446 | 0.7853 | 1600 | 1.0832 |
116
+ | 0.9844 | 0.7975 | 1625 | 1.0833 |
117
+ | 0.9641 | 0.8098 | 1650 | 1.0837 |
118
+ | 1.0 | 0.8221 | 1675 | 1.0835 |
119
+ | 0.9514 | 0.8344 | 1700 | 1.0837 |
120
+ | 1.0094 | 0.8466 | 1725 | 1.0835 |
121
+ | 0.9379 | 0.8589 | 1750 | 1.0834 |
122
+ | 0.9617 | 0.8712 | 1775 | 1.0835 |
123
+ | 0.9674 | 0.8834 | 1800 | 1.0836 |
124
+ | 0.9867 | 0.8957 | 1825 | 1.0838 |
125
+ | 0.9442 | 0.9080 | 1850 | 1.0832 |
126
+ | 0.9603 | 0.9202 | 1875 | 1.0838 |
127
+ | 0.9766 | 0.9325 | 1900 | 1.0833 |
128
+ | 0.9806 | 0.9448 | 1925 | 1.0835 |
129
+ | 0.9676 | 0.9571 | 1950 | 1.0835 |
130
+ | 0.9856 | 0.9693 | 1975 | 1.0838 |
131
+ | 0.9339 | 0.9816 | 2000 | 1.0836 |
132
+ | 0.9553 | 0.9939 | 2025 | 1.0833 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
133
 
134
 
135
  ### Framework versions
136
 
137
+ - Transformers 4.44.2
138
+ - Pytorch 2.0.1+cu117
 
139
  - Tokenizers 0.19.1
generation_config.json CHANGED
@@ -1,4 +1,4 @@
1
  {
2
  "num_beams": 3,
3
- "transformers_version": "4.42.0"
4
  }
 
1
  {
2
  "num_beams": 3,
3
+ "transformers_version": "4.44.2"
4
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:64f9bcac28ca87041eb9c091f8e9317aaab192e67f506f0a60abf315a35a2119
3
  size 1646021682
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5ef8ea56dd4e0b50f3e040b402348281ad3d1acfc49ce45a645e28222489f97
3
  size 1646021682