Felix Marty
commited on
Commit
·
2596ac4
1
Parent(s):
97d3676
fix image display
Browse files
README.md
CHANGED
@@ -16,15 +16,15 @@ To load this model:
|
|
16 |
|
17 |
```python
|
18 |
from optimum.onnxruntime import ORTModelForSequenceClassification
|
19 |
-
model = ORTModelForSequenceClassification.from_pretrained("fxmarty/distilbert-base-uncased-finetuned-sst-2-english-int8-static")
|
20 |
```
|
21 |
|
22 |
<details>
|
23 |
<summary>Weights stored as int8, only DequantizeLinear nodes (model here: https://huggingface.co/fxmarty/distilbert-base-uncased-finetuned-sst-2-english-int8-static)</summary>
|
24 |
-
![DQ only](
|
25 |
</details>
|
26 |
|
27 |
<details>
|
28 |
<summary>Weights stored as fp32, only QuantizeLinear + DequantizeLinear nodes (this model)</summary>
|
29 |
-
![QDQ](
|
30 |
</details>
|
|
|
16 |
|
17 |
```python
|
18 |
from optimum.onnxruntime import ORTModelForSequenceClassification
|
19 |
+
model = ORTModelForSequenceClassification.from_pretrained("fxmarty/distilbert-base-uncased-finetuned-sst-2-english-int8-static-dedicated-qdq-everywhere")
|
20 |
```
|
21 |
|
22 |
<details>
|
23 |
<summary>Weights stored as int8, only DequantizeLinear nodes (model here: https://huggingface.co/fxmarty/distilbert-base-uncased-finetuned-sst-2-english-int8-static)</summary>
|
24 |
+
![DQ only](no_qdq.png)
|
25 |
</details>
|
26 |
|
27 |
<details>
|
28 |
<summary>Weights stored as fp32, only QuantizeLinear + DequantizeLinear nodes (this model)</summary>
|
29 |
+
![QDQ](qdq.png)
|
30 |
</details>
|