Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
```python
|
2 |
+
import torch
|
3 |
+
from transformers import AutoTokenizer
|
4 |
+
from transformers import AutoModelForSeq2SeqLM
|
5 |
+
|
6 |
+
tokenizer = AutoTokenizer.from_pretrained("josh-oo/custom-decoder-ats")
|
7 |
+
|
8 |
+
##gerpt
|
9 |
+
|
10 |
+
#model = AutoModelForSeq2SeqLM.from_pretrained("josh-oo/custom-decoder-ats", trust_remote_code=True, revision="35197269f0235992fcc6b8363ca4f48558b624ff")
|
11 |
+
#decoder_tokenizer = AutoTokenizer.from_pretrained("josh-oo/gerpt2")
|
12 |
+
|
13 |
+
##dbmdz
|
14 |
+
|
15 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("josh-oo/custom-decoder-ats", trust_remote_code=True, revision="4accedbe0b57d342d95ff546b6bbd3321451d504")
|
16 |
+
decoder_tokenizer = AutoTokenizer.from_pretrained("josh-oo/german-gpt2-easy")
|
17 |
+
decoder_tokenizer.add_tokens(['<</s>>','<<s>>','<<pad>>'])
|
18 |
+
|
19 |
+
##
|
20 |
+
|
21 |
+
example_text = "In tausenden Schweizer Privathaushalten kümmern sich Haushaltsangestellte um die Wäsche, betreuen die Kinder und sorgen für Sauberkeit. Durchschnittlich bekommen sie für die Arbeit rund 30 Franken pro Stunde Bruttolohn. Der grösste Teil von ihnen erhält aber 28 Franken."
|
22 |
+
|
23 |
+
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
|
24 |
+
model.to(device)
|
25 |
+
model.eval()
|
26 |
+
|
27 |
+
test_input = tokenizer([example_text], return_tensors="pt", padding=True, pad_to_multiple_of=1024)
|
28 |
+
for key, value in test_input.items():
|
29 |
+
test_input[key] = value.to(device)
|
30 |
+
|
31 |
+
outputs = model.generate(**test_input, num_beams=3, max_length=1024)
|
32 |
+
decoder_tokenizer.batch_decode(outputs)
|
33 |
+
```
|