Instructions to use james-burgess/miewid with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use james-burgess/miewid with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="james-burgess/miewid", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("james-burgess/miewid", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Could you please confirm if this weight is the same as that of conservationxlabs/miewid-msv3?
Could you please confirm if this weight is the same as that of conservationxlabs/miewid-msv3?
Yes. extremely close to bitwise parity.
Proven In this HF Space
Comparison
Metric Value
Cosine similarity 1.0000000417
Agreement embeddings are identical
L2 norm (ONNX) 62.748695
L2 norm (PyTorch) 62.748703
L2 norm diff 0.00000763
Max absolute diff 0.00000811
Mean absolute diff 0.00000152
ONNX time 1020.3 ms
PyTorch time 1340.2 ms
@james-burgess You don't have permission to repost the model as MIT License. Conservation X Labs did not post this model as MIT license. Why would you create a new HuggingFace page for our work?