vincentclaes/emoji-predictor
Viewer β’ Updated β’ 14.4k β’ 87
How to use vincentclaes/emoji-predictor with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("zero-shot-image-classification", model="vincentclaes/emoji-predictor")
pipe(
"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png",
candidate_labels=["animals", "humans", "landscape"],
) # Load model directly
from transformers import AutoProcessor, AutoModelForZeroShotImageClassification
processor = AutoProcessor.from_pretrained("vincentclaes/emoji-predictor")
model = AutoModelForZeroShotImageClassification.from_pretrained("vincentclaes/emoji-predictor")This model is Open AI's CLIP, fine-tuned on a few samples.
Try it here: https://huggingface.co/spaces/vincentclaes/emoji-predictor
Precision for predictions and suggestions for a range of samples per emoji:
| Samples | Prediction | Suggestion |
|--------- |------------ |------------ |
| 0 | 0.13 | 0.33 |
| 1 | 0.11 | 0.30 |
| 5 | 0.14 | 0.38 |
| 10 | 0.20 | 0.45 |
| 15 | 0.22 | 0.51 |
| 20 | 0.19 | 0.49 |
| 25 | 0.24 | 0.54 |
| 50 | 0.23 | 0.53 |
| 100 | 0.25 | 0.57 |
| 250 | 0.29 | 0.62 |
| 500 | 0.29 | 0.63 |