Hugging Face's logo

MoonstoneF
/
kosm-checkpoint

Image-to-Text
Transformers
Safetensors
kosmos-2
Generated from Trainer
Model card Files
xet
Community
MoonstoneF commited on
Commit
e2fa974
·
verified ·
1 Parent(s): acd6dc4

Trained on ALL, resize images to 224*224.

Browse files
Files changed (3) hide show
  1. README.md +7 -13
  2. model-00001-of-00002.safetensors +1 -1
  3. model-00002-of-00002.safetensors +1 -1
README.md CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [microsoft/kosmos-2-patch14-224](https://huggingface.co/microsoft/kosmos-2-patch14-224) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.0873
19
 
20
  ## Model description
21
 
@@ -38,8 +38,8 @@ The following hyperparameters were used during training:
38
  - train_batch_size: 2
39
  - eval_batch_size: 2
40
  - seed: 42
41
- - gradient_accumulation_steps: 2
42
- - total_train_batch_size: 4
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_ratio: 0.1
@@ -49,16 +49,10 @@ The following hyperparameters were used during training:
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:------:|:----:|:---------------:|
52
- | 0.1715 | 0.0160 | 100 | 0.1546 |
53
- | 0.1624 | 0.0319 | 200 | 0.1357 |
54
- | 0.1415 | 0.0479 | 300 | 0.1256 |
55
- | 0.1295 | 0.0638 | 400 | 0.1157 |
56
- | 0.1226 | 0.0798 | 500 | 0.1068 |
57
- | 0.1119 | 0.0957 | 600 | 0.1007 |
58
- | 0.1067 | 0.1117 | 700 | 0.0980 |
59
- | 0.1032 | 0.1276 | 800 | 0.0914 |
60
- | 0.099 | 0.1436 | 900 | 0.0882 |
61
- | 0.0965 | 0.1596 | 1000 | 0.0873 |
62
 
63
 
64
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [microsoft/kosmos-2-patch14-224](https://huggingface.co/microsoft/kosmos-2-patch14-224) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.1082
19
 
20
  ## Model description
21
 
 
38
  - train_batch_size: 2
39
  - eval_batch_size: 2
40
  - seed: 42
41
+ - gradient_accumulation_steps: 4
42
+ - total_train_batch_size: 8
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_ratio: 0.1
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:------:|:----:|:---------------:|
52
+ | 0.8821 | 0.0698 | 250 | 0.1363 |
53
+ | 0.132 | 0.1396 | 500 | 0.1163 |
54
+ | 0.1212 | 0.2094 | 750 | 0.1113 |
55
+ | 0.1166 | 0.2792 | 1000 | 0.1082 |
 
 
 
 
 
 
56
 
57
 
58
  ### Framework versions
model-00001-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4932181b0cce09fc255d8412690b1771822e380f2167945124aaa57312ff3070
3
  size 4999738624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ff1fdf17eae2403ee4d9f013b34b351a35f1a100d2b4e580688fd283f2f363b
3
  size 4999738624
model-00002-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:20e10a9551a9a505a2df622decc92c2f6c23328adda1c64ad6c5cd92fd033f6c
3
  size 1658313704
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87c5268dd141074ff1b6395d0649fb54cdf724bb978efdd18e56487dd22e5b63
3
  size 1658313704