Commit
•
28d2cd7
1
Parent(s):
a762645
update readme
Browse files
README.md
CHANGED
|
@@ -14,9 +14,9 @@ base_model:
|
|
| 14 |
pipeline_tag: text-generation
|
| 15 |
---
|
| 16 |
|
| 17 |
-
# 🎉 Swin-DistilBERTimbau
|
| 18 |
|
| 19 |
-
|
| 20 |
at resolution 224x224 and max sequence length of 512 tokens.
|
| 21 |
|
| 22 |
|
|
@@ -29,6 +29,9 @@ The encoder checkpoints come from Swin Trasnformer version pre-trained on ImageN
|
|
| 29 |
The code used for training and evaluation is available at: https://github.com/laicsiifes/ved-transformer-caption-ptbr. In this work, Swin-DistilBERTimbau
|
| 30 |
was trained together with its buddy [Swin-GPorTuguese](https://huggingface.co/laicsiifes/swin-gpt2-flickr30k-pt-br).
|
| 31 |
|
|
|
|
|
|
|
|
|
|
| 32 |
## 🧑💻 How to Get Started with the Model
|
| 33 |
|
| 34 |
Use the code below to get started with the model.
|
|
|
|
| 14 |
pipeline_tag: text-generation
|
| 15 |
---
|
| 16 |
|
| 17 |
+
# 🎉 Swin-DistilBERTimbau for Image Captioning
|
| 18 |
|
| 19 |
+
Swin-DistilBERTimbau model trained for image captioning on [Flickr30K Portuguese](https://huggingface.co/datasets/laicsiifes/flickr30k-pt-br) (translated version using Google Translator API)
|
| 20 |
at resolution 224x224 and max sequence length of 512 tokens.
|
| 21 |
|
| 22 |
|
|
|
|
| 29 |
The code used for training and evaluation is available at: https://github.com/laicsiifes/ved-transformer-caption-ptbr. In this work, Swin-DistilBERTimbau
|
| 30 |
was trained together with its buddy [Swin-GPorTuguese](https://huggingface.co/laicsiifes/swin-gpt2-flickr30k-pt-br).
|
| 31 |
|
| 32 |
+
Other models evaluated didn't achieve performance as high as Swin-DistilBERTimbau and Swin-GPorTuguese, namely: DeiT-BERTimbau,
|
| 33 |
+
DeiT-DistilBERTimbau, DeiT-GPorTuguese, Swin-BERTimbau, ViT-BERTimbau, ViT-DistilBERTimbau and ViT-GPorTuguese.
|
| 34 |
+
|
| 35 |
## 🧑💻 How to Get Started with the Model
|
| 36 |
|
| 37 |
Use the code below to get started with the model.
|