First version of Donut DR Matriculas

README.md

@@ -1,3 +1,180 @@
----
-license: mit
----
+# Donut 🍩 for DR Matriculas (Donut-DR-matriculas-OCR)
+
+---
+license: mit
+inference: false
+base_model: naver-clova-ix/donut-base
+tags:
+- donut
+- image-to-text
+- vision
+model-index:
+- name: donut-dr-matriculas-ocr
+  results:
+  - task:
+      type: image-to-text
+      name: Image to text
+    metrics:
+    - type: loss
+      value: 0.0563
+      name: Final loss (50 epochs)
+    - type: accuracy
+      value: 0.724689
+      name: F1 Accuracy (Val)
+    - type: accuracy
+      value: 0.923603
+      name: F1 Accuracy (Train)
+    - type: edit distance
+      value: 0.914544
+      name: ED (Val)
+    - type: edit distance
+      value: 0.971895
+      name: ED (Train)      
+metrics:
+- accuracy
+datasets:
+- propietary/matriculas
+pipeline_tag: image-to-text
+---
+
+# Donut-DR-matriculas-OCR
+
+Donut model was introduced in the paper [OCR-free Document Understanding Transformer](https://arxiv.org/abs/2111.15664) by Geewok et al. and first released in [this repository](https://github.com/clovaai/donut).
+
+## === Matriculas OCR V1 ===
+
+This model is a finetune of the [donut base model](https://huggingface.co/naver-clova-ix/donut-base/) on a propietary dataset. Its purpose is to efficiently extract text from the dominican official vehicle registration documents.
+
+This propietary dataset was manually corrected, and we prepared the teacher forcing (ground truth) data with the images and json lines. The license for the V1 model is **mit**, available under the MIT license.
+
+It achieves the following results on the evaluation set:
+
+* Loss: 0.0563
+* Edit distance: 0.914544
+* F1 accuracy: 0.724689
+
+The task_prompt has been changed to ``<s_matricula>`` for the V1.
+
+The focus for the next or future version, will be to collect a better an larger dataset for training.
+
+## Model description
+
+Donut consists of a vision encoder (Swin Transformer) and a text decoder (BART). Given an image, the encoder first encodes the image into a tensor of embeddings (of shape batch_size, seq_len, hidden_size), after which the decoder autoregressively generates text, conditioned on the encoding of the encoder. 
+
+![model image](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/donut_architecture.jpg)
+
+
+### How to use
+
+```python
+import torch
+import re
+from PIL import Image
+from transformers import DonutProcessor
+#from transformers import VisionEncoderDecoderModel
+
+import warnings
+warnings.filterwarnings("ignore")
+
+from sconf import Config
+from donut import DonutConfig, DonutModel
+
+config = Config(default="./config.yaml")
+
+device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
+processor = DonutProcessor.from_pretrained("marzanconsulting/donut-dr-matriculas-ocr")
+
+model = DonutModel.from_pretrained(
+                "marzanconsulting/donut-dr-matriculas-ocr",
+                input_size=config.input_size,
+                max_length=config.max_length,
+                align_long_axis=config.align_long_axis,
+                ignore_mismatched_sizes=True,
+            )
+
+model.to(device)
+
+def load_and_preprocess_image(image_path: str, processor):
+    """
+    Load an image and preprocess it for the model.
+    """
+    image = Image.open(image_path).convert("RGB")
+    pixel_values = processor(image, return_tensors="pt").pixel_values
+    return pixel_values
+
+def generate_text_from_image(model, image_path: str, processor, device):
+    """
+    Generate text from an image using the trained model.
+    """
+    # Load and preprocess the image
+    pixel_values = load_and_preprocess_image(image_path, processor)
+    pixel_values = pixel_values.to(device)
+
+    decoder_input_ids = processor.tokenizer(task_prompt="<s_matricula>", 
+                                            add_special_tokens=False,
+                                            return_tensors="pt").input_ids    
+
+    decoded_text = model.inference(image_tensors=pixel_values, 
+                                   prompt_tensors=decoder_input_ids)["predictions"][0]
+
+    return decoded_text
+
+# Example usage
+image_path = "path_to_your_image"  # Replace with your image path
+extracted_text = generate_text_from_image(model, image_path, processor, device)
+print("Extracted Text:", extracted_text)
+```
+
+Refer to the [documentation](https://huggingface.co/docs/transformers/main/en/model_doc/donut) for more code examples.
+
+## Intended uses & limitations
+
+This fine-tuned model is specifically designed for extracting text from dominican vehicle registration (matriculas) documents, and may not perform optimally on other types of documents. The dataset used is still suboptimal (numerous errors are still there), thus, this model will need to be retrained later to improve its performance.
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 3e-05
+- train_batch_size: 5
+- eval_batch_size: 1
+- seed: 2022
+- optimizer: AdamW with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 300
+- num_epochs: 50
+- weight_decay: 0.01
+
+### Framework versions
+
+- Transformers 4.25.1
+- Timm 0.6.13
+- Pytorch-lightning 1.6.4
+- Donut 1.0.9
+
+If you want to support me, you can [here](https://www.marzanconsulting.com/).
+
+### BibTeX entry and citation info for DONUT
+
+```bibtex
+@article{DBLP:journals/corr/abs-2111-15664,
+  author    = {Geewook Kim and
+               Teakgyu Hong and
+               Moonbin Yim and
+               Jinyoung Park and
+               Jinyeong Yim and
+               Wonseok Hwang and
+               Sangdoo Yun and
+               Dongyoon Han and
+               Seunghyun Park},
+  title     = {Donut: Document Understanding Transformer without {OCR}},
+  journal   = {CoRR},
+  volume    = {abs/2111.15664},
+  year      = {2021},
+  url       = {https://arxiv.org/abs/2111.15664},
+  eprinttype = {arXiv},
+  eprint    = {2111.15664},
+  timestamp = {Thu, 02 Dec 2021 10:50:44 +0100},
+  biburl    = {https://dblp.org/rec/journals/corr/abs-2111-15664.bib},
+  bibsource = {dblp computer science bibliography, https://dblp.org}
+}
+```

added_tokens.json

@@ -0,0 +1,69 @@
+{
+  "</s_ano_de_fabricacion>": 57553,
+  "</s_cap_carga_ton>": 57565,
+  "</s_cedula_pasaporte_rnc>": 57573,
+  "</s_chasis>": 57539,
+  "</s_cilindros>": 57567,
+  "</s_color>": 57557,
+  "</s_datos_del_propietario>": 57571,
+  "</s_datos_del_vehiculo>": 57533,
+  "</s_datos_dgii>": 57581,
+  "</s_dgii_dato_1>": 57585,
+  "</s_dgii_dato_2>": 57583,
+  "</s_direccion>": 57579,
+  "</s_encabezado>": 57527,
+  "</s_estatus_del_vehiculo>": 57541,
+  "</s_fecha_de_expedicion>": 57529,
+  "</s_fields>": 57525,
+  "</s_fuerza_motriz_hp_cc>": 57563,
+  "</s_marca>": 57549,
+  "</s_modelo>": 57551,
+  "</s_motor_o_no_de_serie>": 57559,
+  "</s_no_de_registro_y_placa>": 57537,
+  "</s_no_de_registro_y_placa_anterior>": 57577,
+  "</s_no_matricula>": 57531,
+  "</s_no_puertas>": 57569,
+  "</s_nombres_razon_social>": 57575,
+  "</s_pasajeros>": 57561,
+  "</s_r1>": 57535,
+  "</s_r2>": 57545,
+  "</s_r3>": 57555,
+  "</s_tipo_de_emision>": 57543,
+  "</s_tipo_de_vehiculo>": 57547,
+  "<s_ano_de_fabricacion>": 57554,
+  "<s_cap_carga_ton>": 57566,
+  "<s_cedula_pasaporte_rnc>": 57574,
+  "<s_chasis>": 57540,
+  "<s_cilindros>": 57568,
+  "<s_color>": 57558,
+  "<s_dataset>": 57587,
+  "<s_datos_del_propietario>": 57572,
+  "<s_datos_del_vehiculo>": 57534,
+  "<s_datos_dgii>": 57582,
+  "<s_dgii_dato_1>": 57586,
+  "<s_dgii_dato_2>": 57584,
+  "<s_direccion>": 57580,
+  "<s_encabezado>": 57528,
+  "<s_estatus_del_vehiculo>": 57542,
+  "<s_fecha_de_expedicion>": 57530,
+  "<s_fields>": 57526,
+  "<s_fuerza_motriz_hp_cc>": 57564,
+  "<s_iitcdip>": 57523,
+  "<s_marca>": 57550,
+  "<s_matricula>": 57588,
+  "<s_modelo>": 57552,
+  "<s_motor_o_no_de_serie>": 57560,
+  "<s_no_de_registro_y_placa>": 57538,
+  "<s_no_de_registro_y_placa_anterior>": 57578,
+  "<s_no_matricula>": 57532,
+  "<s_no_puertas>": 57570,
+  "<s_nombres_razon_social>": 57576,
+  "<s_pasajeros>": 57562,
+  "<s_r1>": 57536,
+  "<s_r2>": 57546,
+  "<s_r3>": 57556,
+  "<s_synthdog>": 57524,
+  "<s_tipo_de_emision>": 57544,
+  "<s_tipo_de_vehiculo>": 57548,
+  "<sep/>": 57522
+}

config.json

@@ -0,0 +1,24 @@
+{
+  "_name_or_path": "naver-clova-ix/donut-base",
+  "align_long_axis": false,
+  "architectures": [
+    "DonutModel"
+  ],
+  "decoder_layer": 4,
+  "encoder_layer": [
+    2,
+    2,
+    14,
+    2
+  ],
+  "input_size": [
+    960,
+    1280
+  ],
+  "max_length": 868,
+  "max_position_embeddings": 868,
+  "model_type": "donut",
+  "torch_dtype": "float32",
+  "transformers_version": "4.25.1",
+  "window_size": 10
+}

config.yaml

@@ -0,0 +1,35 @@
+resume_from_checkpoint_path: None
+result_path: 'trainer/result'
+pretrained_model_name_or_path: 'marzanconsulting/donut-dr-matriculas-ocr'
+dataset_name_or_paths: 
+  - 'trainer/dataset'
+task_start_tokens: 
+  - '<s_matricula>'
+sort_json_key: False
+train_batch_sizes: 
+  - 5
+val_batch_sizes: 
+  - 1
+input_size: 
+  - 960
+  - 1280
+max_length: 868
+max_position_embeddings: 868
+align_long_axis: False
+num_nodes: 1
+seed: 2022
+lr: 3e-05
+warmup_steps: 300
+num_training_samples_per_epoch: 800
+max_epochs: 50
+max_steps: -1
+num_workers: 12
+val_check_interval: 1.0
+check_val_every_n_epoch: 3
+gradient_clip_val: 1.0
+verbose: True
+model_dir: ''
+tensorboard_dir: ''
+checkpoint_dir: 
+exp_name: 'train_cord_matriculas'
+exp_version: 'mmc_v1'

preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_align_long_axis": false,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_resize": true,
+  "do_thumbnail": true,
+  "image_processor_type": "DonutImageProcessor",
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "DonutProcessor",
+  "resample": 2,
+  "size": {
+    "width": 1280,
+    "height": 960
+  }
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c16610f407d1f75997923ab15f57662352f256a01688a8a76531a87749c5408a
+size 859005897

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb9e3dce4c326195d08fc3dd0f7e2eee1da8595c847bf4c1a9c78b7a82d47e2d
+size 1296245

special_tokens_map.json

@@ -0,0 +1,19 @@
+{
+  "additional_special_tokens": [
+    "<s_dataset>",
+    "<s_matricula>"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "from_slow": true,
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "naver-clova-ix/donut-base",
+  "pad_token": "<pad>",
+  "processor_class": "DonutProcessor",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "special_tokens_map_file": null,
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}