Image-to-Text
Transformers
Safetensors
Azerbaijani
glm_ocr
image-text-to-text
ocr
vision-language
markdown
azerbaijani
Instructions to use orucexe/azocr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use orucexe/azocr 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="orucexe/azocr")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForMultimodalLM tokenizer = AutoTokenizer.from_pretrained("orucexe/azocr") model = AutoModelForMultimodalLM.from_pretrained("orucexe/azocr", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# AZOCR
|
| 2 |
+
|
| 3 |
+
Azerbaijani OCR model based on GLM-OCR.
|
| 4 |
+
|
| 5 |
+
## Training
|
| 6 |
+
|
| 7 |
+
- Base model: GLM-OCR
|
| 8 |
+
- Language: Azerbaijani
|
| 9 |
+
- Dataset: 10,237 samples
|
| 10 |
+
- Method: LoRA
|
| 11 |
+
- GPUs: 2× NVIDIA Tesla T4
|
| 12 |
+
- Epochs: 2
|
| 13 |
+
|
| 14 |
+
## Notes
|
| 15 |
+
|
| 16 |
+
Fine-tuned for Azerbaijani document OCR.
|