|
--- |
|
library_name: transformers |
|
license: gemma |
|
base_model: google/paligemma-3b-pt-224 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: paligemma_racer |
|
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_racer |
|
|
|
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: 2.9411 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 16 |
|
- optimizer: Use 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: 100 |
|
- num_epochs: 2 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 13.5721 | 0.0209 | 50 | 7.2124 | |
|
| 5.9601 | 0.0419 | 100 | 5.0726 | |
|
| 4.731 | 0.0628 | 150 | 4.4045 | |
|
| 4.2565 | 0.0837 | 200 | 4.0800 | |
|
| 4.0418 | 0.1047 | 250 | 3.8702 | |
|
| 3.866 | 0.1256 | 300 | 3.7370 | |
|
| 3.6864 | 0.1465 | 350 | 3.5834 | |
|
| 3.649 | 0.1675 | 400 | 3.5023 | |
|
| 3.572 | 0.1884 | 450 | 3.4903 | |
|
| 3.4765 | 0.2093 | 500 | 3.4307 | |
|
| 3.406 | 0.2302 | 550 | 3.3801 | |
|
| 3.3997 | 0.2512 | 600 | 3.3027 | |
|
| 3.3602 | 0.2721 | 650 | 3.2871 | |
|
| 3.2852 | 0.2930 | 700 | 3.2509 | |
|
| 3.3183 | 0.3140 | 750 | 3.2354 | |
|
| 3.3281 | 0.3349 | 800 | 3.2133 | |
|
| 3.2545 | 0.3558 | 850 | 3.2098 | |
|
| 3.3173 | 0.3768 | 900 | 3.1909 | |
|
| 3.1993 | 0.3977 | 950 | 3.1646 | |
|
| 3.1705 | 0.4186 | 1000 | 3.1401 | |
|
| 3.1976 | 0.4396 | 1050 | 3.1217 | |
|
| 3.1514 | 0.4605 | 1100 | 3.1340 | |
|
| 3.1832 | 0.4814 | 1150 | 3.1282 | |
|
| 3.1222 | 0.5024 | 1200 | 3.0997 | |
|
| 3.1003 | 0.5233 | 1250 | 3.0788 | |
|
| 3.0833 | 0.5442 | 1300 | 3.0735 | |
|
| 3.099 | 0.5651 | 1350 | 3.0665 | |
|
| 3.1295 | 0.5861 | 1400 | 3.0534 | |
|
| 3.0962 | 0.6070 | 1450 | 3.0392 | |
|
| 3.0589 | 0.6279 | 1500 | 3.0325 | |
|
| 3.075 | 0.6489 | 1550 | 3.0311 | |
|
| 3.034 | 0.6698 | 1600 | 3.0461 | |
|
| 3.0333 | 0.6907 | 1650 | 3.0190 | |
|
| 3.0494 | 0.7117 | 1700 | 3.0174 | |
|
| 3.071 | 0.7326 | 1750 | 3.0123 | |
|
| 3.0147 | 0.7535 | 1800 | 3.0020 | |
|
| 3.0114 | 0.7745 | 1850 | 3.0074 | |
|
| 3.0635 | 0.7954 | 1900 | 3.0224 | |
|
| 2.9939 | 0.8163 | 1950 | 2.9942 | |
|
| 3.0373 | 0.8373 | 2000 | 2.9888 | |
|
| 2.998 | 0.8582 | 2050 | 2.9905 | |
|
| 3.0004 | 0.8791 | 2100 | 2.9883 | |
|
| 2.9477 | 0.9001 | 2150 | 2.9887 | |
|
| 2.9837 | 0.9210 | 2200 | 2.9830 | |
|
| 2.9501 | 0.9419 | 2250 | 2.9788 | |
|
| 3.0235 | 0.9628 | 2300 | 2.9877 | |
|
| 3.0083 | 0.9838 | 2350 | 2.9723 | |
|
| 2.9368 | 1.0047 | 2400 | 2.9775 | |
|
| 2.9975 | 1.0256 | 2450 | 2.9712 | |
|
| 2.9089 | 1.0466 | 2500 | 2.9616 | |
|
| 2.9285 | 1.0675 | 2550 | 2.9669 | |
|
| 2.9627 | 1.0884 | 2600 | 2.9668 | |
|
| 2.9195 | 1.1094 | 2650 | 2.9683 | |
|
| 2.9319 | 1.1303 | 2700 | 2.9607 | |
|
| 2.9009 | 1.1512 | 2750 | 2.9592 | |
|
| 2.9486 | 1.1722 | 2800 | 2.9525 | |
|
| 2.9416 | 1.1931 | 2850 | 2.9532 | |
|
| 2.9223 | 1.2140 | 2900 | 2.9547 | |
|
| 2.9257 | 1.2350 | 2950 | 2.9520 | |
|
| 2.9182 | 1.2559 | 3000 | 2.9516 | |
|
| 2.9255 | 1.2768 | 3050 | 2.9502 | |
|
| 2.9113 | 1.2977 | 3100 | 2.9579 | |
|
| 2.9165 | 1.3187 | 3150 | 2.9584 | |
|
| 2.8901 | 1.3396 | 3200 | 2.9528 | |
|
| 2.921 | 1.3605 | 3250 | 2.9470 | |
|
| 2.9299 | 1.3815 | 3300 | 2.9481 | |
|
| 2.9728 | 1.4024 | 3350 | 2.9458 | |
|
| 2.919 | 1.4233 | 3400 | 2.9446 | |
|
| 2.9132 | 1.4443 | 3450 | 2.9446 | |
|
| 2.9178 | 1.4652 | 3500 | 2.9486 | |
|
| 2.9293 | 1.4861 | 3550 | 2.9450 | |
|
| 2.9514 | 1.5071 | 3600 | 2.9431 | |
|
| 2.9099 | 1.5280 | 3650 | 2.9444 | |
|
| 2.9292 | 1.5489 | 3700 | 2.9449 | |
|
| 2.9336 | 1.5699 | 3750 | 2.9445 | |
|
| 2.8772 | 1.5908 | 3800 | 2.9446 | |
|
| 2.9389 | 1.6117 | 3850 | 2.9444 | |
|
| 2.9618 | 1.6327 | 3900 | 2.9448 | |
|
| 2.9721 | 1.6536 | 3950 | 2.9425 | |
|
| 2.9052 | 1.6745 | 4000 | 2.9406 | |
|
| 2.9245 | 1.6954 | 4050 | 2.9448 | |
|
| 2.9196 | 1.7164 | 4100 | 2.9429 | |
|
| 2.9622 | 1.7373 | 4150 | 2.9408 | |
|
| 2.9199 | 1.7582 | 4200 | 2.9394 | |
|
| 2.9114 | 1.7792 | 4250 | 2.9385 | |
|
| 2.9548 | 1.8001 | 4300 | 2.9402 | |
|
| 2.9263 | 1.8210 | 4350 | 2.9405 | |
|
| 2.9079 | 1.8420 | 4400 | 2.9414 | |
|
| 2.9144 | 1.8629 | 4450 | 2.9367 | |
|
| 2.8985 | 1.8838 | 4500 | 2.9412 | |
|
| 2.8942 | 1.9048 | 4550 | 2.9446 | |
|
| 2.91 | 1.9257 | 4600 | 2.9424 | |
|
| 2.8951 | 1.9466 | 4650 | 2.9414 | |
|
| 2.9054 | 1.9676 | 4700 | 2.9411 | |
|
| 2.8909 | 1.9885 | 4750 | 2.9411 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.3 |
|
- Pytorch 2.5.1 |
|
- Datasets 3.1.0 |
|
- Tokenizers 0.20.3 |
|
|
