Image-Text-to-Text
PEFT
Safetensors
Indonesian
English
fine-tuned
receipts
berkerdooo
conversational
Instructions to use berkerdooo/gemma-4-e2b-receipt-json with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use berkerdooo/gemma-4-e2b-receipt-json with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("google/gemma-4-E2B-it") model = PeftModel.from_pretrained(base_model, "berkerdooo/gemma-4-e2b-receipt-json") - Notebooks
- Google Colab
- Kaggle
Download processor_settings.json from berkerdooo/gemma-4-e2b-receipt-json: direct link, hf CLI and curl.
- Browser
- Download file 388 Bytes
-
https://huggingface.co/berkerdooo/gemma-4-e2b-receipt-json/resolve/main/processor_settings.json
- Command line
-
hf download hf://berkerdooo/gemma-4-e2b-receipt-json/processor_settings.json
-
curl -L -o processor_settings.json https://huggingface.co/berkerdooo/gemma-4-e2b-receipt-json/resolve/main/processor_settings.json
388 Bytes
| { | |
| "actual_max_soft_tokens": 280, | |
| "explicit_default_reproduces_original_tensors": true, | |
| "tensor_shapes": { | |
| "input_ids": [ | |
| 1, | |
| 275 | |
| ], | |
| "attention_mask": [ | |
| 1, | |
| 275 | |
| ], | |
| "mm_token_type_ids": [ | |
| 1, | |
| 275 | |
| ], | |
| "pixel_values": [ | |
| 1, | |
| 2520, | |
| 768 | |
| ], | |
| "image_position_ids": [ | |
| 1, | |
| 2520, | |
| 2 | |
| ] | |
| } | |
| } |