io-roboto's picture
GQA[:100000]
c693f14
---
license: apache-2.0
base_model: dandelin/vilt-b32-mlm
tags:
- generated_from_trainer
model-index:
- name: vilt_finetuned_100000
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. -->
# vilt_finetuned_100000
This model is a fine-tuned version of [dandelin/vilt-b32-mlm](https://huggingface.co/dandelin/vilt-b32-mlm) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2049
## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 148.4684 | 0.04 | 250 | 5.6972 |
| 5.0473 | 0.08 | 500 | 4.5986 |
| 4.4685 | 0.12 | 750 | 4.2115 |
| 4.0423 | 0.16 | 1000 | 3.8991 |
| 3.8019 | 0.2 | 1250 | 3.7080 |
| 3.6731 | 0.24 | 1500 | 3.5319 |
| 3.5226 | 0.28 | 1750 | 3.4277 |
| 3.3556 | 0.32 | 2000 | 3.3546 |
| 3.279 | 0.36 | 2250 | 3.2592 |
| 3.2406 | 0.4 | 2500 | 3.1792 |
| 3.1275 | 0.44 | 2750 | 3.1014 |
| 3.0569 | 0.48 | 3000 | 3.0938 |
| 3.0484 | 0.52 | 3250 | 2.9939 |
| 2.9883 | 0.56 | 3500 | 2.9230 |
| 2.9351 | 0.6 | 3750 | 2.8902 |
| 2.8526 | 0.64 | 4000 | 2.8242 |
| 2.8469 | 0.68 | 4250 | 2.8069 |
| 2.7228 | 0.72 | 4500 | 2.7304 |
| 2.6549 | 0.76 | 4750 | 2.6875 |
| 2.6304 | 0.8 | 5000 | 2.6461 |
| 2.6239 | 0.84 | 5250 | 2.6117 |
| 2.6334 | 0.88 | 5500 | 2.5914 |
| 2.622 | 0.92 | 5750 | 2.5320 |
| 2.5581 | 0.96 | 6000 | 2.5028 |
| 2.536 | 1.0 | 6250 | 2.5663 |
| 2.3565 | 1.04 | 6500 | 2.4770 |
| 2.2907 | 1.08 | 6750 | 2.4965 |
| 2.3328 | 1.12 | 7000 | 2.4514 |
| 2.2924 | 1.16 | 7250 | 2.4503 |
| 2.2799 | 1.2 | 7500 | 2.4083 |
| 2.2477 | 1.24 | 7750 | 2.4246 |
| 2.2662 | 1.28 | 8000 | 2.3792 |
| 2.2656 | 1.32 | 8250 | 2.3568 |
| 2.1932 | 1.36 | 8500 | 2.3682 |
| 2.1821 | 1.4 | 8750 | 2.3241 |
| 2.1663 | 1.44 | 9000 | 2.3182 |
| 2.1017 | 1.48 | 9250 | 2.3225 |
| 2.1884 | 1.52 | 9500 | 2.3069 |
| 2.1873 | 1.56 | 9750 | 2.2613 |
| 2.1094 | 1.6 | 10000 | 2.2730 |
| 2.0764 | 1.64 | 10250 | 2.2694 |
| 2.1051 | 1.68 | 10500 | 2.2537 |
| 2.1283 | 1.72 | 10750 | 2.2434 |
| 2.1212 | 1.76 | 11000 | 2.2164 |
| 2.0978 | 1.8 | 11250 | 2.2168 |
| 1.9994 | 1.84 | 11500 | 2.2015 |
| 2.0101 | 1.88 | 11750 | 2.1916 |
| 2.0156 | 1.92 | 12000 | 2.1699 |
| 2.014 | 1.96 | 12250 | 2.1821 |
| 1.9576 | 2.0 | 12500 | 2.1739 |
| 1.6838 | 2.04 | 12750 | 2.2527 |
| 1.6692 | 2.08 | 13000 | 2.2432 |
| 1.6237 | 2.12 | 13250 | 2.2824 |
| 1.5776 | 2.16 | 13500 | 2.2619 |
| 1.6468 | 2.2 | 13750 | 2.2499 |
| 1.6279 | 2.24 | 14000 | 2.2454 |
| 1.6277 | 2.28 | 14250 | 2.2624 |
| 1.6293 | 2.32 | 14500 | 2.2283 |
| 1.6217 | 2.36 | 14750 | 2.2720 |
| 1.5921 | 2.4 | 15000 | 2.2643 |
| 1.578 | 2.44 | 15250 | 2.2336 |
| 1.5559 | 2.48 | 15500 | 2.2311 |
| 1.5932 | 2.52 | 15750 | 2.2063 |
| 1.5532 | 2.56 | 16000 | 2.2205 |
| 1.5736 | 2.6 | 16250 | 2.2141 |
| 1.5931 | 2.64 | 16500 | 2.2399 |
| 1.5735 | 2.68 | 16750 | 2.2106 |
| 1.5161 | 2.72 | 17000 | 2.1973 |
| 1.5388 | 2.76 | 17250 | 2.2009 |
| 1.5139 | 2.8 | 17500 | 2.2083 |
| 1.5298 | 2.84 | 17750 | 2.2030 |
| 1.5339 | 2.88 | 18000 | 2.2031 |
| 1.4774 | 2.92 | 18250 | 2.2050 |
| 1.5067 | 2.96 | 18500 | 2.2074 |
| 1.4847 | 3.0 | 18750 | 2.2049 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1