File size: 1,587 Bytes
32043f4 0976156 698c218 0976156 32043f4 0976156 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
---
pipeline_tag: image-classification
---
## Table cell classification
The model is trained to classify table cell images as either empty or not empty. It has been trained using
table cell images from Finnish census and death record tables from the 1930s.
The model has been trained using [densenet121](https://pytorch.org/vision/stable/models/generated/torchvision.models.densenet121.html) as the base model,
and it has been transformed into the [onnx](https://onnx.ai/) format.
## Intended uses & limitations
The model has been trained to classify table cells from specific kinds of tables, which contain mainly handwritten text.
It has not been tested with other type of table cell data.
## Training and validation data
Training dataset consisted of
- empty cell images: 2943
- non-empty cell images: 5033
Validation dataset consisted of
- empty cell images: 367
- non-empty cell images: 627
## Training procedure
The code used for model training is available in the repository in `train.py` file, which uses functions from
`augment.py` and `utils.py` files. The required libraries are listed in the `requirements.txt` file.
The model was trained using cpu with the following hyperparameters:
- image size: 2560
- learning rate: 0.0001
- train batch size: 32
- epochs: 15
- patience: 3 epochs
- optimizer: Adam
## Evaluation results
Evaluation results using the validation dataset are listed below:
|Validation loss|Validation accuracy|Validation F1-score
-|-|-
0.0427|0.9899|0.9903
## Inference
Inference can be performed using the code in the `test.py` file. |