Image-to-Text
Transformers
Safetensors
Portuguese
vision-encoder-decoder
image-text-to-text
Eval Results (legacy)
Instructions to use laicsiifes/swin-distilbertimbau with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use laicsiifes/swin-distilbertimbau with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="laicsiifes/swin-distilbertimbau")# Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("laicsiifes/swin-distilbertimbau") model = AutoModelForMultimodalLM.from_pretrained("laicsiifes/swin-distilbertimbau", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -9,8 +9,34 @@ metrics:
|
|
| 9 |
- rouge
|
| 10 |
- meteor
|
| 11 |
- bertscore
|
| 12 |
-
base_model: laicsiifes/swin-
|
| 13 |
pipeline_tag: image-to-text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
---
|
| 15 |
|
| 16 |
# 🎉 Swin-DistilBERTimbau for Brazilian Portuguese Image Captioning
|
|
|
|
| 9 |
- rouge
|
| 10 |
- meteor
|
| 11 |
- bertscore
|
| 12 |
+
base_model: laicsiifes/swin-distilbertimbau
|
| 13 |
pipeline_tag: image-to-text
|
| 14 |
+
model-index:
|
| 15 |
+
- name: Swin-DistilBERTimbau
|
| 16 |
+
results:
|
| 17 |
+
- task:
|
| 18 |
+
name: Image Captioning
|
| 19 |
+
type: image-to-text
|
| 20 |
+
dataset:
|
| 21 |
+
name: ai2_arc
|
| 22 |
+
type: ai2_arc
|
| 23 |
+
metrics:
|
| 24 |
+
- name: Cider-D
|
| 25 |
+
type: cider
|
| 26 |
+
value: 66.73%
|
| 27 |
+
- name: BLEU@4
|
| 28 |
+
type: bleu
|
| 29 |
+
value: 24.65%
|
| 30 |
+
- name: ROUGE-L
|
| 31 |
+
type: rouge
|
| 32 |
+
value: 39.98%
|
| 33 |
+
- name: METEOR
|
| 34 |
+
type: meteor
|
| 35 |
+
value: 44.71%
|
| 36 |
+
- name: BERTScore
|
| 37 |
+
type: bertscore
|
| 38 |
+
value: 72.30%
|
| 39 |
+
|
| 40 |
---
|
| 41 |
|
| 42 |
# 🎉 Swin-DistilBERTimbau for Brazilian Portuguese Image Captioning
|