Training in progress, step 500

README.md

@@ -1,33 +1,54 @@
 ---
 license: cc-by-nc-sa-4.0
-base_model: microsoft/layoutxlm-base
 tags:
 - generated_from_trainer
+base_model: microsoft/layoutxlm-base
 datasets:
-- xfun
+- nielsr/XFUN
+inference: false
 model-index:
 - name: layoutxlm-finetuned-xfund-fr
   results: []
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
 # layoutxlm-finetuned-xfund-fr
 
-This model is a fine-tuned version of [microsoft/layoutxlm-base](https://huggingface.co/microsoft/layoutxlm-base) on the xfun dataset.
+This model is a fine-tuned version of [microsoft/layoutxlm-base](https://huggingface.co/microsoft/layoutxlm-base) on the [XFUND](https://github.com/doc-analysis/XFUND) dataset (French split).
+
+## Model usage
+
+Note that this model requires Tesseract, French package, in order to perform inference. You can install it using `!sudo apt-get install tesseract-ocr-fra`.
+
+Here's how to use this model:
+
+```
+from transformers import AutoProcessor, AutoModelForTokenClassification
+import torch
+from PIL import Image
+
+processor = AutoProcessor.from_pretrained("nielsr/layoutxlm-finetuned-xfund-fr")
+model = AutoModelForTokenClassification.from_pretrained(nielsr/layoutxlm-finetuned-xfund-fr")
+
+# assuming you have a French document, turned into an image
+image = Image.open("...").convert("RGB")
 
-## Model description
+# prepare for the model
+encoding = processor(image, padding="max_length", max_length=512, truncation=True, return_tensors="pt")
 
-More information needed
+with torch.no_grad():
+  outputs = model(**encoding)
+  logits = outputs.logits
+  
+predictions = logits.argmax(-1)
+```
 
 ## Intended uses & limitations
 
-More information needed
+This model can be used for NER on French scanned documents. It can recognize 4 categories: "question", "answer", "header" and "other".
 
 ## Training and evaluation data
 
-More information needed
+This checkpoint used the French portion of the multilingual [XFUND](https://github.com/doc-analysis/XFUND) dataset.
 
 ## Training procedure
 
@@ -49,7 +70,7 @@ The following hyperparameters were used during training:
 
 ### Framework versions
 
-- Transformers 4.40.2
-- Pytorch 2.2.1+cu121
-- Datasets 2.19.1
-- Tokenizers 0.19.1
+- Transformers 4.22.1
+- Pytorch 1.10.0+cu111
+- Datasets 2.4.0
+- Tokenizers 0.12.1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f9caadd0bdc338a48dcaa52bfa97398371329e4b151a4fb0726e282f68b4514
+oid sha256:794f4a557585b8c4d08e144cf7904852b9dec64dd129c925d48b41469b5ded43
 size 1476349244

runs/May16_08-28-48_7299b205dec4/events.out.tfevents.1715848129.7299b205dec4.2577.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fddf4c37859f7fcfd004b114b302dfd70f849b185323b9fe116672801680c4b
+size 7332

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be0fdd1000f135ccdb957b04d9b32eeded260e8f1d0f5d96567b7c3c42bf7019
+oid sha256:ee8943a370dea5344c32308e18461cad7049538d7f24d25d4df109876375a86c
 size 4984