Hugging Face's logo

minishlab
/
potion-base-8m-onnx

Model2Vec
ONNX
sentence-transformers
embeddings
static-embeddings
Model card Files
xet
 Community
stephantulkens commited on
Commit
e893cf4
·
verified ·
1 Parent(s): a7114e7

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. README.md +3 -3
  2. model.onnx +2 -2
  3. tokenizer_config.json +1 -1
README.md CHANGED
@@ -2,7 +2,7 @@
2
  base_model: minishlab/potion-base-8m
3
  library_name: model2vec
4
  license: mit
5
- model_name: potion-base-8m-onnx
6
  tags:
7
  - embeddings
8
  - static-embeddings
@@ -10,7 +10,7 @@ tags:
10
  - onnx
11
  ---
12
 
13
- # potion-base-8m-onnx Model Card
14
 
15
  This is an ONNX export of the [minishlab/potion-base-8m](https://huggingface.co/minishlab/potion-base-8m) [Model2Vec](https://github.com/MinishLab/model2vec) model, produced with the [ONNX](https://onnx.ai/) runtime. [Model2Vec](https://github.com/MinishLab/model2vec) models use static embeddings, allowing text embeddings to be computed orders of magnitude faster on both GPU and CPU. This ONNX export lets you run the model with `onnxruntime` or `transformers.js`, without depending on the `model2vec` package.
16
 
@@ -22,7 +22,7 @@ This is an ONNX export of the [minishlab/potion-base-8m](https://huggingface.co/
22
  import onnxruntime as ort
23
  from transformers import AutoTokenizer
24
 
25
- tokenizer = AutoTokenizer.from_pretrained("potion-base-8m-onnx")
26
  session = ort.InferenceSession("model.onnx")
27
 
28
  encodings = tokenizer(["Example sentence"], padding=True, return_tensors="np")
 
2
  base_model: minishlab/potion-base-8m
3
  library_name: model2vec
4
  license: mit
5
+ model_name: potion-base-8m_onnx
6
  tags:
7
  - embeddings
8
  - static-embeddings
 
10
  - onnx
11
  ---
12
 
13
+ # potion-base-8m_onnx Model Card
14
 
15
  This is an ONNX export of the [minishlab/potion-base-8m](https://huggingface.co/minishlab/potion-base-8m) [Model2Vec](https://github.com/MinishLab/model2vec) model, produced with the [ONNX](https://onnx.ai/) runtime. [Model2Vec](https://github.com/MinishLab/model2vec) models use static embeddings, allowing text embeddings to be computed orders of magnitude faster on both GPU and CPU. This ONNX export lets you run the model with `onnxruntime` or `transformers.js`, without depending on the `model2vec` package.
16
 
 
22
  import onnxruntime as ort
23
  from transformers import AutoTokenizer
24
 
25
+ tokenizer = AutoTokenizer.from_pretrained("potion-base-8m_onnx")
26
  session = ort.InferenceSession("model.onnx")
27
 
28
  encodings = tokenizer(["Example sentence"], padding=True, return_tensors="np")
model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d24d36c6d43600067012301c4b12dfa29d50c5acbae3f9e5a3d2de8bad6a0fb5
3
- size 30248074
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b955aeda5bb8eaedb3c302dc3757ad3bcbe2efc4e1b6091be21c28b3a3f392eb
3
+ size 30251506
tokenizer_config.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "backend": "tokenizers",
3
  "clean_up_tokenization_spaces": false,
4
- "model_max_length": 32768,
5
  "pad_to_multiple_of": null,
6
  "pad_token": "[PAD]",
7
  "pad_token_type_id": 0,
 
1
  {
2
  "backend": "tokenizers",
3
  "clean_up_tokenization_spaces": false,
4
+ "model_max_length": 512,
5
  "pad_to_multiple_of": null,
6
  "pad_token": "[PAD]",
7
  "pad_token_type_id": 0,