File size: 3,068 Bytes
a603253
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
740fc92
 
a603253
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f691844
a603253
 
 
 
 
 
 
 
538bfdb
a603253
 
 
740fc92
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a603253
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
---
library_name: peft
license: gemma
base_model: google/paligemma-3b-pt-224
tags:
- generated_from_trainer
model-index:
- name: paligemma_vqav2
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# paligemma_vqav2

This model is a fine-tuned version of [google/paligemma-3b-pt-224](https://huggingface.co/google/paligemma-3b-pt-224) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2568

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_HF with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 10

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 6.2845        | 0.3137 | 100  | 1.7300          |
| 0.9773        | 0.6275 | 200  | 0.6286          |
| 0.5181        | 0.9412 | 300  | 0.4802          |
| 0.4126        | 1.2549 | 400  | 0.4208          |
| 0.365         | 1.5686 | 500  | 0.3774          |
| 0.3369        | 1.8824 | 600  | 0.3645          |
| 0.3037        | 2.1961 | 700  | 0.3299          |
| 0.2854        | 2.5098 | 800  | 0.3164          |
| 0.2939        | 2.8235 | 900  | 0.3093          |
| 0.2547        | 3.1373 | 1000 | 0.2961          |
| 0.2275        | 3.4510 | 1100 | 0.2943          |
| 0.2456        | 3.7647 | 1200 | 0.2824          |
| 0.2368        | 4.0784 | 1300 | 0.2723          |
| 0.2148        | 4.3922 | 1400 | 0.2733          |
| 0.2118        | 4.7059 | 1500 | 0.2737          |
| 0.1991        | 5.0196 | 1600 | 0.2715          |
| 0.1879        | 5.3333 | 1700 | 0.2657          |
| 0.1841        | 5.6471 | 1800 | 0.2746          |
| 0.1912        | 5.9608 | 1900 | 0.2642          |
| 0.1509        | 6.2745 | 2000 | 0.2964          |
| 0.1818        | 6.5882 | 2100 | 0.2607          |
| 0.1736        | 6.9020 | 2200 | 0.2644          |
| 0.1618        | 7.2157 | 2300 | 0.2663          |
| 0.1563        | 7.5294 | 2400 | 0.2637          |
| 0.159         | 7.8431 | 2500 | 0.2561          |
| 0.1488        | 8.1569 | 2600 | 0.2554          |
| 0.1417        | 8.4706 | 2700 | 0.2589          |
| 0.1329        | 8.7843 | 2800 | 0.2599          |
| 0.1455        | 9.0980 | 2900 | 0.2589          |
| 0.1521        | 9.4118 | 3000 | 0.2576          |
| 0.1334        | 9.7255 | 3100 | 0.2568          |


### Framework versions

- PEFT 0.13.2
- Transformers 4.46.2
- Pytorch 2.4.0.post301
- Datasets 3.1.0
- Tokenizers 0.20.3