Zero-Shot Image Classification
OpenCLIP
ONNX
English
clip
vision
aerial
drone
tracking
re-identification
Instructions to use llama-farm/llama-thunderdome-clip-aerial-vit-b16-v3-drone-fleet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use llama-farm/llama-thunderdome-clip-aerial-vit-b16-v3-drone-fleet with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:llama-farm/llama-thunderdome-clip-aerial-vit-b16-v3-drone-fleet') tokenizer = open_clip.get_tokenizer('hf-hub:llama-farm/llama-thunderdome-clip-aerial-vit-b16-v3-drone-fleet') - Notebooks
- Google Colab
- Kaggle
Upload validate_report.json with huggingface_hub
Browse files- validate_report.json +15 -0
validate_report.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"verdict": "PASS",
|
| 3 |
+
"n_samples": 20,
|
| 4 |
+
"cosine": {
|
| 5 |
+
"mean": 0.9550989866256714,
|
| 6 |
+
"std": 0.01038367673754692,
|
| 7 |
+
"min": 0.9314883947372437,
|
| 8 |
+
"max": 0.9719838500022888
|
| 9 |
+
},
|
| 10 |
+
"overlap_at_5": 0.8200000000000001,
|
| 11 |
+
"threshold": 0.85,
|
| 12 |
+
"preprocess": "letterbox uint8 [0,255], norm1 in HEF graph",
|
| 13 |
+
"quant_har": "/tmp/v3-export/hef/clip_aerial_vit_b_16_quantized.har",
|
| 14 |
+
"reference_pt": "/tmp/v3-export/best.pt"
|
| 15 |
+
}
|