Update model to overflowfix version
Browse files
README.md
CHANGED
@@ -5,30 +5,16 @@ tags:
|
|
5 |
datasets:
|
6 |
- squad
|
7 |
model-index:
|
8 |
-
- name:
|
9 |
results: []
|
10 |
---
|
11 |
|
12 |
-
|
13 |
-
should probably proofread and complete it, then remove this comment. -->
|
14 |
-
|
15 |
-
# jpqd_bert_squad_overflowfix
|
16 |
|
17 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the squad dataset.
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
More information needed
|
22 |
-
|
23 |
-
## Intended uses & limitations
|
24 |
-
|
25 |
-
More information needed
|
26 |
-
|
27 |
-
## Training and evaluation data
|
28 |
-
|
29 |
-
More information needed
|
30 |
-
|
31 |
-
## Training procedure
|
32 |
|
33 |
### Training hyperparameters
|
34 |
|
@@ -42,9 +28,6 @@ The following hyperparameters were used during training:
|
|
42 |
- num_epochs: 8.0
|
43 |
- mixed_precision_training: Native AMP
|
44 |
|
45 |
-
### Training results
|
46 |
-
|
47 |
-
|
48 |
|
49 |
### Framework versions
|
50 |
|
@@ -52,3 +35,4 @@ The following hyperparameters were used during training:
|
|
52 |
- Pytorch 1.13.1+cu117
|
53 |
- Datasets 2.8.0
|
54 |
- Tokenizers 0.13.2
|
|
|
|
5 |
datasets:
|
6 |
- squad
|
7 |
model-index:
|
8 |
+
- name: bert-base-uncased-squad-v1-jpqd-ov-int8
|
9 |
results: []
|
10 |
---
|
11 |
|
12 |
+
# bert-base-uncased-squad-v1-jpqd-ov-int8
|
|
|
|
|
|
|
13 |
|
14 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the squad dataset.
|
15 |
+
It was compressed using [NNCF](https://github.com/openvinotoolkit/nncf) with [Optimum
|
16 |
+
Intel](https://github.com/huggingface/optimum-intel#openvino) following the [JPQD question-answering
|
17 |
+
example](https://github.com/huggingface/optimum-intel/tree/main/examples/openvino/question-answering#joint-pruning-quantization-and-distillation-jpqd-for-bert-on-squad10).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
|
19 |
### Training hyperparameters
|
20 |
|
|
|
28 |
- num_epochs: 8.0
|
29 |
- mixed_precision_training: Native AMP
|
30 |
|
|
|
|
|
|
|
31 |
|
32 |
### Framework versions
|
33 |
|
|
|
35 |
- Pytorch 1.13.1+cu117
|
36 |
- Datasets 2.8.0
|
37 |
- Tokenizers 0.13.2
|
38 |
+
|