Upload folder using huggingface_hub
Browse files
.gitattributes
CHANGED
|
@@ -1,35 +1 @@
|
|
| 1 |
-
*.
|
| 2 |
-
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
-
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
-
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
-
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
-
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
-
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
-
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
-
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
-
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
-
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
-
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
-
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
-
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
-
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
-
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
-
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
-
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
-
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
-
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
-
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
-
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
-
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
-
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
-
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
-
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
-
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
-
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
-
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
-
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
-
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
-
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
-
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
-
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
-
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 1 |
+
*.ort filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
LICENSE
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
MIT License
|
| 2 |
+
|
| 3 |
+
The model weights in model.ort are those of intfloat/multilingual-e5-small
|
| 4 |
+
(https://huggingface.co/intfloat/multilingual-e5-small), revision
|
| 5 |
+
614241f622f53c4eeff9890bdc4f31cfecc418b3, released under the MIT licence
|
| 6 |
+
(`license: mit` in that repository's model card; the repository carries no
|
| 7 |
+
separate LICENSE file at that revision, so the standard MIT licence text
|
| 8 |
+
follows). Converted to the ONNX Runtime (ORT) format by Mentisia.
|
| 9 |
+
|
| 10 |
+
Copyright (c) Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, Furu Wei
|
| 11 |
+
(intfloat, the authors of multilingual-e5-small)
|
| 12 |
+
|
| 13 |
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
| 14 |
+
of this software and associated documentation files (the "Software"), to deal
|
| 15 |
+
in the Software without restriction, including without limitation the rights
|
| 16 |
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
| 17 |
+
copies of the Software, and to permit persons to whom the Software is
|
| 18 |
+
furnished to do so, subject to the following conditions:
|
| 19 |
+
|
| 20 |
+
The above copyright notice and this permission notice shall be included in all
|
| 21 |
+
copies or substantial portions of the Software.
|
| 22 |
+
|
| 23 |
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
| 24 |
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
| 25 |
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
| 26 |
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
| 27 |
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
| 28 |
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
| 29 |
+
SOFTWARE.
|
README.md
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: onnxruntime
|
| 4 |
+
base_model: intfloat/multilingual-e5-small
|
| 5 |
+
tags:
|
| 6 |
+
- onnxruntime
|
| 7 |
+
- ort-format
|
| 8 |
+
- sentence-embeddings
|
| 9 |
+
- e5
|
| 10 |
+
- multilingual
|
| 11 |
+
- ios
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# multilingual-e5-small in ORT format (memory-mappable)
|
| 15 |
+
|
| 16 |
+
The weights of [intfloat/multilingual-e5-small](https://huggingface.co/intfloat/multilingual-e5-small)
|
| 17 |
+
in the [ONNX Runtime ORT format](https://onnxruntime.ai/docs/performance/model-optimizations/ort-format-models.html),
|
| 18 |
+
for **Memoa for iOS**. Loaded with its initialisers memory-mapped, it embeds with a fraction of the
|
| 19 |
+
memory the ONNX file needs, and gives **bit-identical** vectors.
|
| 20 |
+
|
| 21 |
+
| | |
|
| 22 |
+
|---|---|
|
| 23 |
+
| Source | `intfloat/multilingual-e5-small`, revision `614241f622f53c4eeff9890bdc4f31cfecc418b3`, `onnx/model.onnx` (470,268,510 bytes, SHA-256 `ca456c06b3a9505ddfd9131408916dd79290368331e7d76bb621f1cba6bc8665`) |
|
| 24 |
+
| File | `model.ort`: 470,350,304 bytes, SHA-256 `ba3519110cbb648984f04415ebc6e32946de1c55ac85afbe3719864a76b4ef29` |
|
| 25 |
+
| Converter | ONNX Runtime **1.31.0** (Python wheel, installed by hash), macOS arm64 |
|
| 26 |
+
| Session options | CPU execution provider only; intra-op and inter-op threads 2; `GraphOptimizationLevel.ORT_ENABLE_ALL`; `optimized_model_filepath` with `session.save_model_format=ORT` |
|
| 27 |
+
| Script | `scripts/convert-e5-ort.sh` in Memoa for iOS (`mentisia-memoa-ios`) |
|
| 28 |
+
|
| 29 |
+
## What it is
|
| 30 |
+
|
| 31 |
+
It is the same graph ONNX Runtime 1.31.0 builds from the source `model.onnx` at every load with
|
| 32 |
+
these options, saved once. Nothing is retrained, quantised or pruned; the weights are unchanged
|
| 33 |
+
(float32).
|
| 34 |
+
|
| 35 |
+
To load it with the initialisers mapped from the file, use ONNX Runtime 1.31.0 (CPU) and set
|
| 36 |
+
`session.load_model_format=ORT`, `session.use_memory_mapped_ort_model=1` and
|
| 37 |
+
`session.use_ort_model_bytes_for_initializers=1`.
|
| 38 |
+
|
| 39 |
+
ONNX Runtime warns that a model optimised at `ORT_ENABLE_ALL` may hold hardware-specific layout
|
| 40 |
+
transforms (NCHWc, an x86 convolution layout). e5 has no convolutions, and the vectors below were
|
| 41 |
+
checked on arm64 (an M1 Mac and an A16 iPad). Two conversions differ only in about 1 KB of
|
| 42 |
+
flatbuffer metadata, with the same vectors. This file is the one Memoa pins.
|
| 43 |
+
|
| 44 |
+
## Vectors
|
| 45 |
+
|
| 46 |
+
The 33 reference texts of Memoa's e5 parity fixture (25 passages, 8 queries, made with memoryd's
|
| 47 |
+
fastembed / ONNX Runtime 1.31.0 stack) were embedded from this file, with mapped initialisers, and
|
| 48 |
+
from the source `model.onnx`:
|
| 49 |
+
|
| 50 |
+
- every vector is **bit-identical** to the source model's (M1 Mac: this file; iPad A16: a conversion
|
| 51 |
+
of the same source by the same ORT version);
|
| 52 |
+
- max |difference| to the reference vectors 9.69e-08, int8 codes (× 640) identical.
|
| 53 |
+
|
| 54 |
+
On an iPad (A16), the load's peak memory footprint was **264 MB with this file**, against 1,351 MB
|
| 55 |
+
for `model.onnx` read from its path. The load took 0.64 s instead of 1.03–1.23 s.
|
| 56 |
+
|
| 57 |
+
## Attribution and citation
|
| 58 |
+
|
| 59 |
+
The model is by **intfloat** (Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder,
|
| 60 |
+
Furu Wei); see the [original model card](https://huggingface.co/intfloat/multilingual-e5-small)
|
| 61 |
+
for training data, intended use, and limitations, which all apply here unchanged.
|
| 62 |
+
|
| 63 |
+
```
|
| 64 |
+
@article{wang2024multilingual,
|
| 65 |
+
title={Multilingual E5 Text Embeddings: A Technical Report},
|
| 66 |
+
author={Wang, Liang and Yang, Nan and Huang, Xiaolong and Yang, Linjun and Majumder, Rangan and Wei, Furu},
|
| 67 |
+
journal={arXiv preprint arXiv:2402.05672},
|
| 68 |
+
year={2024}
|
| 69 |
+
}
|
| 70 |
+
```
|
| 71 |
+
|
| 72 |
+
As with the original, prefix inputs with `query: ` or `passage: `.
|
| 73 |
+
|
| 74 |
+
Licence: MIT, as the original (see `LICENSE`).
|
model.ort
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba3519110cbb648984f04415ebc6e32946de1c55ac85afbe3719864a76b4ef29
|
| 3 |
+
size 470350304
|