Instructions to use manishw10/devgen-trocr-devanagari-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use manishw10/devgen-trocr-devanagari-lora with PEFT:
from peft import PeftModel from transformers import AutoModelForSeq2SeqLM base_model = AutoModelForSeq2SeqLM.from_pretrained("paudelanil/trocr-devanagari-2") model = PeftModel.from_pretrained(base_model, "manishw10/devgen-trocr-devanagari-lora") - Transformers
How to use manishw10/devgen-trocr-devanagari-lora 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="manishw10/devgen-trocr-devanagari-lora")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("manishw10/devgen-trocr-devanagari-lora", dtype="auto") - Notebooks
- Google Colab
- Kaggle
manishw7 commited on
Commit ·
ecce7a8
1
Parent(s): 6dc6d66
Fix: Restore 1.75 AR threshold for character routing
Browse files
app.py
CHANGED
|
@@ -81,6 +81,7 @@ def original_classify_input(image):
|
|
| 81 |
elif bc >= 4: is_char = False
|
| 82 |
elif ar < 1.3 and bc <= 2: is_char = True
|
| 83 |
elif bc == 1 and ar < 1.5: is_char = True
|
|
|
|
| 84 |
elif ar > 1.6: is_char = False
|
| 85 |
return ("character" if is_char else "word"), ar, bc
|
| 86 |
|
|
|
|
| 81 |
elif bc >= 4: is_char = False
|
| 82 |
elif ar < 1.3 and bc <= 2: is_char = True
|
| 83 |
elif bc == 1 and ar < 1.5: is_char = True
|
| 84 |
+
elif ar < 1.75 and bc <= 2: is_char = True # <--- RESTORED THIS LINE
|
| 85 |
elif ar > 1.6: is_char = False
|
| 86 |
return ("character" if is_char else "word"), ar, bc
|
| 87 |
|