This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +4 -0
  2. README.md +36 -0
  3. onnx/_roberta_encoder_layers.0_mixer_rotary_emb_Constant_6_attr__value +3 -0
  4. onnx/_roberta_encoder_layers.0_mixer_rotary_emb_Constant_attr__value +3 -0
  5. onnx/model.onnx +3 -0
  6. onnx/roberta.embeddings.token_type_embeddings.parametrizations.weight.0.lora_B +0 -0
  7. onnx/roberta.embeddings.word_embeddings.parametrizations.weight.0.lora_A +3 -0
  8. onnx/roberta.embeddings.word_embeddings.parametrizations.weight.0.lora_B +0 -0
  9. onnx/roberta.embeddings.word_embeddings.parametrizations.weight.original +3 -0
  10. onnx/roberta.encoder.layers.0.mixer.Wqkv.bias +0 -0
  11. onnx/roberta.encoder.layers.0.mixer.Wqkv.parametrizations.weight.0.lora_A +0 -0
  12. onnx/roberta.encoder.layers.0.mixer.Wqkv.parametrizations.weight.0.lora_B +0 -0
  13. onnx/roberta.encoder.layers.0.mixer.Wqkv.parametrizations.weight.original +3 -0
  14. onnx/roberta.encoder.layers.0.mixer.out_proj.parametrizations.weight.0.lora_A +0 -0
  15. onnx/roberta.encoder.layers.0.mixer.out_proj.parametrizations.weight.0.lora_B +0 -0
  16. onnx/roberta.encoder.layers.0.mixer.out_proj.parametrizations.weight.original +3 -0
  17. onnx/roberta.encoder.layers.0.mlp.fc1.bias +0 -0
  18. onnx/roberta.encoder.layers.0.mlp.fc1.parametrizations.weight.0.lora_A +0 -0
  19. onnx/roberta.encoder.layers.0.mlp.fc1.parametrizations.weight.0.lora_B +0 -0
  20. onnx/roberta.encoder.layers.0.mlp.fc1.parametrizations.weight.original +3 -0
  21. onnx/roberta.encoder.layers.0.mlp.fc2.parametrizations.weight.0.lora_A +0 -0
  22. onnx/roberta.encoder.layers.0.mlp.fc2.parametrizations.weight.0.lora_B +0 -0
  23. onnx/roberta.encoder.layers.0.mlp.fc2.parametrizations.weight.original +3 -0
  24. onnx/roberta.encoder.layers.1.mixer.Wqkv.bias +0 -0
  25. onnx/roberta.encoder.layers.1.mixer.Wqkv.parametrizations.weight.0.lora_A +0 -0
  26. onnx/roberta.encoder.layers.1.mixer.Wqkv.parametrizations.weight.0.lora_B +0 -0
  27. onnx/roberta.encoder.layers.1.mixer.Wqkv.parametrizations.weight.original +3 -0
  28. onnx/roberta.encoder.layers.1.mixer.out_proj.parametrizations.weight.0.lora_A +0 -0
  29. onnx/roberta.encoder.layers.1.mixer.out_proj.parametrizations.weight.0.lora_B +0 -0
  30. onnx/roberta.encoder.layers.1.mixer.out_proj.parametrizations.weight.original +3 -0
  31. onnx/roberta.encoder.layers.1.mlp.fc1.bias +0 -0
  32. onnx/roberta.encoder.layers.1.mlp.fc1.parametrizations.weight.0.lora_A +0 -0
  33. onnx/roberta.encoder.layers.1.mlp.fc1.parametrizations.weight.0.lora_B +0 -0
  34. onnx/roberta.encoder.layers.1.mlp.fc1.parametrizations.weight.original +3 -0
  35. onnx/roberta.encoder.layers.1.mlp.fc2.parametrizations.weight.0.lora_A +0 -0
  36. onnx/roberta.encoder.layers.1.mlp.fc2.parametrizations.weight.0.lora_B +0 -0
  37. onnx/roberta.encoder.layers.1.mlp.fc2.parametrizations.weight.original +3 -0
  38. onnx/roberta.encoder.layers.10.mixer.Wqkv.bias +0 -0
  39. onnx/roberta.encoder.layers.10.mixer.Wqkv.parametrizations.weight.0.lora_A +0 -0
  40. onnx/roberta.encoder.layers.10.mixer.Wqkv.parametrizations.weight.0.lora_B +0 -0
  41. onnx/roberta.encoder.layers.10.mixer.Wqkv.parametrizations.weight.original +3 -0
  42. onnx/roberta.encoder.layers.10.mixer.out_proj.parametrizations.weight.0.lora_A +0 -0
  43. onnx/roberta.encoder.layers.10.mixer.out_proj.parametrizations.weight.0.lora_B +0 -0
  44. onnx/roberta.encoder.layers.10.mixer.out_proj.parametrizations.weight.original +3 -0
  45. onnx/roberta.encoder.layers.10.mlp.fc1.bias +0 -0
  46. onnx/roberta.encoder.layers.10.mlp.fc1.parametrizations.weight.0.lora_A +0 -0
  47. onnx/roberta.encoder.layers.10.mlp.fc1.parametrizations.weight.0.lora_B +0 -0
  48. onnx/roberta.encoder.layers.10.mlp.fc1.parametrizations.weight.original +3 -0
  49. onnx/roberta.encoder.layers.10.mlp.fc2.parametrizations.weight.0.lora_A +0 -0
  50. onnx/roberta.encoder.layers.10.mlp.fc2.parametrizations.weight.0.lora_B +0 -0
.gitattributes CHANGED
@@ -34,3 +34,7 @@ saved_model/**/* 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
36
  tokenizer.json 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
36
  tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ *.original filter=lfs diff=lfs merge=lfs -text
38
+ onnx/roberta.embeddings.word_embeddings.parametrizations.weight.0.lora_A filter=lfs diff=lfs merge=lfs -text
39
+ onnx/_roberta_encoder_layers.0_mixer_rotary_emb_Constant_6_attr__value filter=lfs diff=lfs merge=lfs -text
40
+ onnx/_roberta_encoder_layers.0_mixer_rotary_emb_Constant_attr__value filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -21683,6 +21683,42 @@ embeddings = model.encode(
21683
  )
21684
  ```
21685
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21686
 
21687
  ## Contact
21688
 
 
21683
  )
21684
  ```
21685
 
21686
+ **<details><summary>ONNX Inference.</summary>**
21687
+ <p>
21688
+
21689
+ You can use ONNX for efficient inference with `jina-embeddings-v3`:
21690
+ ```python
21691
+ import onnxruntime
21692
+ import numpy as np
21693
+ from transformers import AutoTokenizer, PretrainedConfig
21694
+
21695
+ # Load tokenizer and model config
21696
+ tokenizer = AutoTokenizer.from_pretrained('jinaai/jina-embeddings-v3')
21697
+ config = PretrainedConfig.from_pretrained('jinaai/jina-embeddings-v3')
21698
+
21699
+ # Tokenize input
21700
+ input_text = tokenizer('sample text', return_tensors='np')
21701
+
21702
+ # ONNX session
21703
+ model_path = 'jina-embeddings-v3/onnx/model.onnx'
21704
+ session = onnxruntime.InferenceSession(model_path)
21705
+
21706
+ # Prepare inputs for ONNX model
21707
+ task_type = 'text-matching'
21708
+ task_id = np.array(config.lora_adaptations.index(task_type), dtype=np.int64)
21709
+ inputs = {
21710
+ 'input_ids': input_text['input_ids'],
21711
+ 'attention_mask': input_text['attention_mask'],
21712
+ 'task_id': task_id
21713
+ }
21714
+
21715
+ # Run model
21716
+ outputs = session.run(None, inputs)
21717
+ ```
21718
+
21719
+ </p>
21720
+ </details>
21721
+
21722
 
21723
  ## Contact
21724
 
onnx/_roberta_encoder_layers.0_mixer_rotary_emb_Constant_6_attr__value ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20472209ecbc9a1cea36feaa7e507db402e879d4fc51ebaf5f3cafefc4a7c4a9
3
+ size 1048832
onnx/_roberta_encoder_layers.0_mixer_rotary_emb_Constant_attr__value ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7249c502306a0d781150401d414af5736ec47bed9f96395e5a3b504159342ab1
3
+ size 1048832
onnx/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8586a2e2981d6ccf9d26104d926b92ca45c8c33a55fd146c62519b59aeaf7d2c
3
+ size 3113411
onnx/roberta.embeddings.token_type_embeddings.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.embeddings.word_embeddings.parametrizations.weight.0.lora_A ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e331476c75a983801498b3a97c6e8599359b5002127a6802f94a11f9b67698b3
3
+ size 20000160
onnx/roberta.embeddings.word_embeddings.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.embeddings.word_embeddings.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:827e52a1bff60a42513a0c2ded3bb3db168c973d473ab1a4e3281ec2060b4d98
3
+ size 1024008192
onnx/roberta.encoder.layers.0.mixer.Wqkv.bias ADDED
Binary file (12.3 kB). View file
 
onnx/roberta.encoder.layers.0.mixer.Wqkv.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.0.mixer.Wqkv.parametrizations.weight.0.lora_B ADDED
Binary file (246 kB). View file
 
onnx/roberta.encoder.layers.0.mixer.Wqkv.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:062b6c3a1067b86638020bcebe33f19e3d0a94d86ca95091fb9f465bf76b5c69
3
+ size 12582912
onnx/roberta.encoder.layers.0.mixer.out_proj.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.0.mixer.out_proj.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.0.mixer.out_proj.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a17c40b5eba6671395bcad41b488730162eb8f26aea126d2de8f76857053adb5
3
+ size 4194304
onnx/roberta.encoder.layers.0.mlp.fc1.bias ADDED
Binary file (16.4 kB). View file
 
onnx/roberta.encoder.layers.0.mlp.fc1.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.0.mlp.fc1.parametrizations.weight.0.lora_B ADDED
Binary file (328 kB). View file
 
onnx/roberta.encoder.layers.0.mlp.fc1.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:841cf3f67fbb99cad22eb25d469b1300e1246dc62ae7e651ef7a50c37386098e
3
+ size 16777216
onnx/roberta.encoder.layers.0.mlp.fc2.parametrizations.weight.0.lora_A ADDED
Binary file (328 kB). View file
 
onnx/roberta.encoder.layers.0.mlp.fc2.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.0.mlp.fc2.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a27c74c0d67d8a96ea66a48ca382d4a0a3ebf5354b1ce0bd75a1c8ef2f815ee
3
+ size 16777216
onnx/roberta.encoder.layers.1.mixer.Wqkv.bias ADDED
Binary file (12.3 kB). View file
 
onnx/roberta.encoder.layers.1.mixer.Wqkv.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.1.mixer.Wqkv.parametrizations.weight.0.lora_B ADDED
Binary file (246 kB). View file
 
onnx/roberta.encoder.layers.1.mixer.Wqkv.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1dbe254f0ec39a45b80c711a392141e8d4ff1d54e1aafb0c917855cf6b46612
3
+ size 12582912
onnx/roberta.encoder.layers.1.mixer.out_proj.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.1.mixer.out_proj.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.1.mixer.out_proj.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1af212a115f6af2f0cd3d6430c71d53ea022cda186ed9097fcda06d243471598
3
+ size 4194304
onnx/roberta.encoder.layers.1.mlp.fc1.bias ADDED
Binary file (16.4 kB). View file
 
onnx/roberta.encoder.layers.1.mlp.fc1.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.1.mlp.fc1.parametrizations.weight.0.lora_B ADDED
Binary file (328 kB). View file
 
onnx/roberta.encoder.layers.1.mlp.fc1.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:083cd922bbe4bb93359461a08ca3ab1c80feb9ba08ac9332f07a2f204549fe0f
3
+ size 16777216
onnx/roberta.encoder.layers.1.mlp.fc2.parametrizations.weight.0.lora_A ADDED
Binary file (328 kB). View file
 
onnx/roberta.encoder.layers.1.mlp.fc2.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.1.mlp.fc2.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:004cdb4408206f4a5c0d9c2e0a829ec0096c5ad98e9bbe93af8c0104af0d9e4a
3
+ size 16777216
onnx/roberta.encoder.layers.10.mixer.Wqkv.bias ADDED
Binary file (12.3 kB). View file
 
onnx/roberta.encoder.layers.10.mixer.Wqkv.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.10.mixer.Wqkv.parametrizations.weight.0.lora_B ADDED
Binary file (246 kB). View file
 
onnx/roberta.encoder.layers.10.mixer.Wqkv.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a0013b2bf24bdd46723a059c3bde4eb7347d9d1f894459ba42050209e6ecc7d
3
+ size 12582912
onnx/roberta.encoder.layers.10.mixer.out_proj.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.10.mixer.out_proj.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.10.mixer.out_proj.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bb88e776197f7986730c10af22bbaac3de27496471e5e1a93ba46fe6f3e1d94
3
+ size 4194304
onnx/roberta.encoder.layers.10.mlp.fc1.bias ADDED
Binary file (16.4 kB). View file
 
onnx/roberta.encoder.layers.10.mlp.fc1.parametrizations.weight.0.lora_A ADDED
Binary file (81.9 kB). View file
 
onnx/roberta.encoder.layers.10.mlp.fc1.parametrizations.weight.0.lora_B ADDED
Binary file (328 kB). View file
 
onnx/roberta.encoder.layers.10.mlp.fc1.parametrizations.weight.original ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bcfc8c48fa5682fab4eda4efc926296c1b912e25e8eabbf088826f79cfba627
3
+ size 16777216
onnx/roberta.encoder.layers.10.mlp.fc2.parametrizations.weight.0.lora_A ADDED
Binary file (328 kB). View file
 
onnx/roberta.encoder.layers.10.mlp.fc2.parametrizations.weight.0.lora_B ADDED
Binary file (81.9 kB). View file