File size: 3,491 Bytes
5f801b8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
93
94
95
96
97
98
99
---
base_model: google/paligemma-3b-pt-224
library_name: peft
license: gemma
tags:
- generated_from_trainer
model-index:
- name: paligemma_vqav2_warnup
  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_warnup

This model is a fine-tuned version of [google/paligemma-3b-pt-224](https://huggingface.co/google/paligemma-3b-pt-224) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8534

## 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: 2e-05
- train_batch_size: 10
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 40
- 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: 1000
- num_epochs: 12

### Training results

| Training Loss | Epoch   | Step | Validation Loss |
|:-------------:|:-------:|:----:|:---------------:|
| 7.0325        | 0.3125  | 100  | 7.1622          |
| 6.5146        | 0.625   | 200  | 5.6221          |
| 4.4912        | 0.9375  | 300  | 3.4696          |
| 3.2824        | 1.25    | 400  | 2.6386          |
| 2.6828        | 1.5625  | 500  | 2.1185          |
| 2.2276        | 1.875   | 600  | 1.7020          |
| 1.9293        | 2.1875  | 700  | 1.4613          |
| 1.7957        | 2.5     | 800  | 1.3059          |
| 1.6281        | 2.8125  | 900  | 1.1815          |
| 1.4253        | 3.125   | 1000 | 1.1008          |
| 1.4404        | 3.4375  | 1100 | 1.0513          |
| 1.4239        | 3.75    | 1200 | 1.0156          |
| 1.2985        | 4.0625  | 1300 | 0.9709          |
| 1.243         | 4.375   | 1400 | 0.9393          |
| 1.2105        | 4.6875  | 1500 | 0.9237          |
| 1.2035        | 5.0     | 1600 | 0.9088          |
| 1.0704        | 5.3125  | 1700 | 0.8927          |
| 1.0891        | 5.625   | 1800 | 0.8735          |
| 1.0861        | 5.9375  | 1900 | 0.8598          |
| 0.9853        | 6.25    | 2000 | 0.8530          |
| 0.9866        | 6.5625  | 2100 | 0.8392          |
| 1.0206        | 6.875   | 2200 | 0.8399          |
| 0.8914        | 7.1875  | 2300 | 0.8293          |
| 0.9062        | 7.5     | 2400 | 0.8325          |
| 0.8579        | 7.8125  | 2500 | 0.8147          |
| 0.828         | 8.125   | 2600 | 0.8267          |
| 0.7969        | 8.4375  | 2700 | 0.8321          |
| 0.8175        | 8.75    | 2800 | 0.8179          |
| 0.7948        | 9.0625  | 2900 | 0.8356          |
| 0.7221        | 9.375   | 3000 | 0.8104          |
| 0.7124        | 9.6875  | 3100 | 0.8266          |
| 0.7199        | 10.0    | 3200 | 0.8143          |
| 0.6601        | 10.3125 | 3300 | 0.8399          |
| 0.6517        | 10.625  | 3400 | 0.8415          |
| 0.6509        | 10.9375 | 3500 | 0.8221          |
| 0.5981        | 11.25   | 3600 | 0.8582          |
| 0.6011        | 11.5625 | 3700 | 0.8481          |
| 0.5756        | 11.875  | 3800 | 0.8534          |


### Framework versions

- PEFT 0.13.0
- Transformers 4.46.0.dev0
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.0