Update README.md
Browse files
README.md
CHANGED
|
@@ -41,13 +41,18 @@ pip install "torch>=2.1" "nncf>=2.7" "transformers>=4.40.0" "onnx<1.16.2" "optim
|
|
| 41 |
```
|
| 42 |
This command will install, among others:
|
| 43 |
```
|
| 44 |
-
tokenizers
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
openvino
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
```
|
| 52 |
|
| 53 |
|
|
@@ -143,6 +148,25 @@ console.print(f'Generated in {delta}')
|
|
| 143 |
```
|
| 144 |
|
| 145 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
## Benchmark
|
| 147 |
|
| 148 |
Zero-shot performance (English):
|
|
|
|
| 41 |
```
|
| 42 |
This command will install, among others:
|
| 43 |
```
|
| 44 |
+
tokenizers==0.20.3
|
| 45 |
+
torch==2.5.1+cpu
|
| 46 |
+
transformers==4.46.3
|
| 47 |
+
nncf==2.14.0
|
| 48 |
+
numpy==2.1.3
|
| 49 |
+
onnx==1.16.1
|
| 50 |
+
openvino==2024.5.0
|
| 51 |
+
openvino-genai==2024.5.0.0
|
| 52 |
+
openvino-telemetry==2024.5.0
|
| 53 |
+
openvino-tokenizers==2024.5.0.0
|
| 54 |
+
optimum==1.23.3
|
| 55 |
+
optimum-intel @ git+https://github.com/huggingface/optimum-intel.git@c454b0000279ac9801302d726fbbbc1152733315
|
| 56 |
```
|
| 57 |
|
| 58 |
|
|
|
|
| 148 |
```
|
| 149 |
|
| 150 |
|
| 151 |
+
<p align="left">
|
| 152 |
+
<img src="https://github.com/fabiomatricardi/NuExtract-1.5-openvino/raw/main/logo_long.png" style="width: 600; height: auto;">
|
| 153 |
+
</p>
|
| 154 |
+
|
| 155 |
+
|
| 156 |
+
## An awesome Streamlit+OpenVINO interface [](https://github.com/openvinotoolkit/awesome-openvino)
|
| 157 |
+
you can find the code in [my official GitHub repository](https://github.com/fabiomatricardi/NuExtract-1.5-openvino)
|
| 158 |
+
|
| 159 |
+
You can clone the repo and use the downloaded files from this Hugging Face Model
|
| 160 |
+
|
| 161 |
+
Running the streamlit app will give this:
|
| 162 |
+
|
| 163 |
+
<p align="left">
|
| 164 |
+
<img src="https://github.com/fabiomatricardi/NuExtract-1.5-openvino/raw/main/streamlitapp.png" style="width: 600; height: auto;">
|
| 165 |
+
</p>
|
| 166 |
+
|
| 167 |
+
---
|
| 168 |
+
|
| 169 |
+
|
| 170 |
## Benchmark
|
| 171 |
|
| 172 |
Zero-shot performance (English):
|