Update README.md
Browse files
README.md
CHANGED
@@ -43,6 +43,8 @@ You can use the raw model for text generation or fine-tune it to a downstream ta
|
|
43 |
|
44 |
Here is how to use the ONNX models of gpt2 to get the features of a given text:
|
45 |
|
|
|
|
|
46 |
```python
|
47 |
from transformers import AutoTokenizer, pipeline
|
48 |
from optimum.onnxruntime import ORTModelForCausalLM
|
@@ -54,3 +56,19 @@ onnx_gen = pipeline("text-generation", model=model, tokenizer=tokenizer)
|
|
54 |
text = "My name is Philipp and I live in Germany."
|
55 |
gen = onnx_gen(text)
|
56 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
Here is how to use the ONNX models of gpt2 to get the features of a given text:
|
45 |
|
46 |
+
Example using transformers.pipelines:
|
47 |
+
|
48 |
```python
|
49 |
from transformers import AutoTokenizer, pipeline
|
50 |
from optimum.onnxruntime import ORTModelForCausalLM
|
|
|
56 |
text = "My name is Philipp and I live in Germany."
|
57 |
gen = onnx_gen(text)
|
58 |
```
|
59 |
+
|
60 |
+
Example of text generation:
|
61 |
+
|
62 |
+
```python
|
63 |
+
from transformers import AutoTokenizer
|
64 |
+
from optimum.onnxruntime import ORTModelForCausalLM
|
65 |
+
import torch
|
66 |
+
|
67 |
+
tokenizer = AutoTokenizer.from_pretrained("optimum/gpt2")
|
68 |
+
model = ORTModelForCausalLM.from_pretrained("optimum/gpt2")
|
69 |
+
|
70 |
+
inputs = tokenizer("My name is Arthur and I live in", return_tensors="pt")
|
71 |
+
|
72 |
+
gen_tokens = model.generate(**inputs,do_sample=True,temperature=0.9, min_length=20,max_length=20)
|
73 |
+
tokenizer.batch_decode(gen_tokens)
|
74 |
+
```
|