Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,9 @@ language:
|
|
| 5 |
tags:
|
| 6 |
- sparse sparsity quantized onnx embeddings int8
|
| 7 |
---
|
|
|
|
|
|
|
|
|
|
| 8 |
This is the sparsified ONNX variant of the [bge-large-en-v1.5](https://huggingface.co/BAAI/bge-large-en-v1.5) embeddings model created with [DeepSparse Optimum](https://github.com/neuralmagic/optimum-deepsparse) for ONNX export/inference pipeline and Neural Magic's [Sparsify](https://github.com/neuralmagic/sparsify) for one-shot quantization (INT8) and unstructured pruning (50%).
|
| 9 |
|
| 10 |
Model achieves 98% accuracy recovery on the STSB validation dataset vs. [dense ONNX variant](https://huggingface.co/zeroshot/bge-large-en-v1.5-dense).
|
|
|
|
| 5 |
tags:
|
| 6 |
- sparse sparsity quantized onnx embeddings int8
|
| 7 |
---
|
| 8 |
+
|
| 9 |
+
# bge-large-en-v1.5-sparse
|
| 10 |
+
|
| 11 |
This is the sparsified ONNX variant of the [bge-large-en-v1.5](https://huggingface.co/BAAI/bge-large-en-v1.5) embeddings model created with [DeepSparse Optimum](https://github.com/neuralmagic/optimum-deepsparse) for ONNX export/inference pipeline and Neural Magic's [Sparsify](https://github.com/neuralmagic/sparsify) for one-shot quantization (INT8) and unstructured pruning (50%).
|
| 12 |
|
| 13 |
Model achieves 98% accuracy recovery on the STSB validation dataset vs. [dense ONNX variant](https://huggingface.co/zeroshot/bge-large-en-v1.5-dense).
|