Update README.md
Browse files
README.md
CHANGED
@@ -67,16 +67,16 @@ The next table shows the results as well as a comparison with other German langu
|
|
67 |
### How to use
|
68 |
|
69 |
The simplest way to use the model is the huggingface transformers pipeline tool.
|
70 |
-
Just initialize the pipeline specifying the task as "zero-shot-classification"
|
71 |
and select "svalabs/gbert-large-zeroshot-nli" as model.
|
72 |
|
73 |
-
The model requires you to specify labels
|
74 |
a sequence (or list of sequences) to classify and a hypothesis template.
|
75 |
In our tests, if the labels comprise only single words,
|
76 |
"In diesem Satz geht es um das Thema {}" performed the best.
|
77 |
|
78 |
However, for multiple words, especially when they combine nouns and verbs,
|
79 |
-
Simple hypothesis such as "Weil" or "Daher" may
|
80 |
|
81 |
Here is an example of how to use the model:
|
82 |
|
|
|
67 |
### How to use
|
68 |
|
69 |
The simplest way to use the model is the huggingface transformers pipeline tool.
|
70 |
+
Just initialize the pipeline specifying the task as "zero-shot-classification"
|
71 |
and select "svalabs/gbert-large-zeroshot-nli" as model.
|
72 |
|
73 |
+
The model requires you to specify labels,
|
74 |
a sequence (or list of sequences) to classify and a hypothesis template.
|
75 |
In our tests, if the labels comprise only single words,
|
76 |
"In diesem Satz geht es um das Thema {}" performed the best.
|
77 |
|
78 |
However, for multiple words, especially when they combine nouns and verbs,
|
79 |
+
Simple hypothesis such as "Weil {}" or "Daher {}" may work better.
|
80 |
|
81 |
Here is an example of how to use the model:
|
82 |
|