|
|
--- |
|
|
library_name: transformers |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
- zh |
|
|
tags: |
|
|
- remote-sensing |
|
|
- mllm |
|
|
- multimodal |
|
|
- earth-observation |
|
|
- satellite-imagery |
|
|
pipeline_tag: image-text-to-text |
|
|
--- |
|
|
|
|
|
# 🌍 TerraSense-Base |
|
|
|
|
|
A Multimodal Large Language Model for Remote Sensing. |
|
|
|
|
|
## 📖 Documentation |
|
|
|
|
|
For usage instructions, examples, and detailed documentation, please visit: |
|
|
|
|
|
👉 **[GitHub Repository](https://github.com/TerraSense-CASM/terrasense)** |
|
|
|
|
|
## 🚀 Quick Start |
|
|
|
|
|
```python |
|
|
from transformers import AutoModelForVision2Seq, AutoProcessor |
|
|
from qwen_vl_utils import process_vision_info |
|
|
import torch |
|
|
|
|
|
model = AutoModelForVision2Seq.from_pretrained( |
|
|
"TerraSense-CASM/TerraSense-Base", |
|
|
torch_dtype=torch.bfloat16, |
|
|
device_map="auto", |
|
|
trust_remote_code=True |
|
|
) |
|
|
processor = AutoProcessor.from_pretrained("TerraSense-CASM/TerraSense-Base", trust_remote_code=True) |
|
|
|
|
|
messages = [{"role": "user", "content": [ |
|
|
{"type": "image", "image": "path/to/image.jpg"}, |
|
|
{"type": "text", "text": "Describe this remote sensing image."}, |
|
|
]}] |
|
|
|
|
|
text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) |
|
|
image_inputs, _ = process_vision_info(messages) |
|
|
inputs = processor(text=[text], images=image_inputs, padding=True, return_tensors="pt").to("cuda") |
|
|
output = model.generate(**inputs, max_new_tokens=512) |
|
|
print(processor.batch_decode(output, skip_special_tokens=True)[0]) |
|
|
``` |
|
|
|
|
|
## 📜 License |
|
|
|
|
|
[Apache 2.0](https://github.com/TerraSense-CASM/terrasense/blob/main/LICENSE) |
|
|
|
