Update README.md
Browse files
README.md
CHANGED
|
@@ -17,6 +17,18 @@ This is BERT trained on [SinglePage DocVQA](https://arxiv.org/abs/2007.00398) an
|
|
| 17 |
This model was used as a baseline in [Hierarchical multimodal transformers for Multi-Page DocVQA](https://arxiv.org/pdf/2212.05935.pdf).
|
| 18 |
- Training hyperparameters can be found in Table 8 of Appendix D.
|
| 19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## How to use
|
| 22 |
|
|
@@ -33,21 +45,7 @@ encoded_input = tokenizer(question, context, return_tensors='pt')
|
|
| 33 |
output = model(**encoded_input)
|
| 34 |
```
|
| 35 |
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
Extended experimentation can be found in Table 2 of [Hierarchical multimodal transformers for Multi-Page DocVQA](https://arxiv.org/pdf/2212.05935.pdf).
|
| 39 |
-
You can also check the live leaderboard at the [RRC Portal](https://rrc.cvc.uab.es/?ch=17&com=evaluation&task=4).
|
| 40 |
-
| Model | HF name | ANLS | APPA |
|
| 41 |
-
|-----------------------------------------------------------------------------------|:--------------------------------------|:-------------:|:---------:|
|
| 42 |
-
| [**Bert-large**](https://huggingface.co/rubentito/bert-large-mpdocvqa/tree/main) | rubentito/bert-large-mpdocvqa | 0.4183 | 51.6177 |
|
| 43 |
-
| [Longformer-base](https://huggingface.co/rubentito/longformer-base-mpdocvqa) | rubentito/longformer-base-mpdocvqa | 0.5287 | 71.1696 |
|
| 44 |
-
| [BigBird ITC base](https://huggingface.co/rubentito/bigbird-base-itc-mpdocvqa) | rubentito/bigbird-base-itc-mpdocvqa | 0.4929 | 67.5433 |
|
| 45 |
-
| [LayoutLMv3 base](https://huggingface.co/rubentito/layoutlmv3-base-mpdocvqa) | rubentito/layoutlmv3-base-mpdocvqa | 0.4538 | 51.9426 |
|
| 46 |
-
| [T5 base](https://huggingface.co/rubentito/t5-base-mpdocvqa) | rubentito/t5-base-mpdocvqa | 0.5050 | 0.0000 |
|
| 47 |
-
| Hi-VT5 | TBA | 0.6201 | 79.23 |
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
## BibTeX entry
|
| 52 |
|
| 53 |
```tex
|
|
|
|
| 17 |
This model was used as a baseline in [Hierarchical multimodal transformers for Multi-Page DocVQA](https://arxiv.org/pdf/2212.05935.pdf).
|
| 18 |
- Training hyperparameters can be found in Table 8 of Appendix D.
|
| 19 |
|
| 20 |
+
## Model results
|
| 21 |
+
|
| 22 |
+
Extended experimentation can be found in Table 2 of [Hierarchical multimodal transformers for Multi-Page DocVQA](https://arxiv.org/pdf/2212.05935.pdf).
|
| 23 |
+
You can also check the live leaderboard at the [RRC Portal](https://rrc.cvc.uab.es/?ch=17&com=evaluation&task=4).
|
| 24 |
+
| Model | HF name | ANLS | APPA |
|
| 25 |
+
|-----------------------------------------------------------------------------------|:--------------------------------------|:-------------:|:---------:|
|
| 26 |
+
| [**Bert-large**](https://huggingface.co/rubentito/bert-large-mpdocvqa) | rubentito/bert-large-mpdocvqa | 0.4183 | 51.6177 |
|
| 27 |
+
| [Longformer-base](https://huggingface.co/rubentito/longformer-base-mpdocvqa) | rubentito/longformer-base-mpdocvqa | 0.5287 | 71.1696 |
|
| 28 |
+
| [BigBird ITC base](https://huggingface.co/rubentito/bigbird-base-itc-mpdocvqa) | rubentito/bigbird-base-itc-mpdocvqa | 0.4929 | 67.5433 |
|
| 29 |
+
| [LayoutLMv3 base](https://huggingface.co/rubentito/layoutlmv3-base-mpdocvqa) | rubentito/layoutlmv3-base-mpdocvqa | 0.4538 | 51.9426 |
|
| 30 |
+
| [T5 base](https://huggingface.co/rubentito/t5-base-mpdocvqa) | rubentito/t5-base-mpdocvqa | 0.5050 | 0.0000 |
|
| 31 |
+
| Hi-VT5
|
| 32 |
|
| 33 |
## How to use
|
| 34 |
|
|
|
|
| 45 |
output = model(**encoded_input)
|
| 46 |
```
|
| 47 |
|
| 48 |
+
| TBA | 0.6201 | 79.23
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
## BibTeX entry
|
| 50 |
|
| 51 |
```tex
|