Update README.md
Browse files
README.md
CHANGED
@@ -27,6 +27,10 @@ This model is designed for NLP tasks involving Quranic text in Bahasa Indonesia,
|
|
27 |
|
28 |
## Uses
|
29 |
|
|
|
|
|
|
|
|
|
30 |
### Downstream Use
|
31 |
|
32 |
It is suitable for fine-tuning on tasks such as:
|
@@ -34,7 +38,6 @@ It is suitable for fine-tuning on tasks such as:
|
|
34 |
- Question answering systems related to Islamic knowledge
|
35 |
- Educational tools for learning Quranic content in Indonesian
|
36 |
|
37 |
-
|
38 |
### Biases
|
39 |
- The model inherits any biases present in the dataset, which is specific to Islamic translations and tafsir in Bahasa Indonesia.
|
40 |
|
@@ -47,10 +50,36 @@ It is suitable for fine-tuning on tasks such as:
|
|
47 |
```python
|
48 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
49 |
|
50 |
-
tokenizer
|
51 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
|
|
27 |
|
28 |
## Uses
|
29 |
|
30 |
+
### Direct Use
|
31 |
+
|
32 |
+
This model can be used for applications requiring the understanding, summarization, or retrieval of Quranic translations and tafsir in Bahasa Indonesia.
|
33 |
+
|
34 |
### Downstream Use
|
35 |
|
36 |
It is suitable for fine-tuning on tasks such as:
|
|
|
38 |
- Question answering systems related to Islamic knowledge
|
39 |
- Educational tools for learning Quranic content in Indonesian
|
40 |
|
|
|
41 |
### Biases
|
42 |
- The model inherits any biases present in the dataset, which is specific to Islamic translations and tafsir in Bahasa Indonesia.
|
43 |
|
|
|
50 |
```python
|
51 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
52 |
|
53 |
+
# Load the tokenizer and model
|
54 |
+
tokenizer = AutoTokenizer.from_pretrained("Ellbendls/Qwen-2.5-3b-Quran")
|
55 |
+
model = AutoModelForCausalLM.from_pretrained("Ellbendls/Qwen-2.5-3b-Quran")
|
56 |
+
|
57 |
+
# Move the model to GPU
|
58 |
+
model.to("cuda")
|
59 |
+
|
60 |
+
# Define the input message
|
61 |
+
messages = [
|
62 |
+
{
|
63 |
+
"role": "user",
|
64 |
+
"content": "Tafsirkan ayat ini اِهْدِنَا الصِّرَاطَ الْمُسْتَقِيْمَۙ"
|
65 |
+
}
|
66 |
+
]
|
67 |
+
|
68 |
+
# Generate the prompt using the tokenizer
|
69 |
+
prompt = tokenizer.apply_chat_template(messages, tokenize=False,
|
70 |
+
add_generation_prompt=True)
|
71 |
+
|
72 |
+
# Tokenize the prompt and move inputs to GPU
|
73 |
+
inputs = tokenizer(prompt, return_tensors='pt', padding=True,
|
74 |
+
truncation=True).to("cuda")
|
75 |
+
|
76 |
+
# Generate the output using the model
|
77 |
+
outputs = model.generate(**inputs, max_length=150,
|
78 |
+
num_return_sequences=1)
|
79 |
+
|
80 |
+
# Decode the output
|
81 |
+
text = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
82 |
|
83 |
+
# Print the result
|
84 |
+
print(text.split("assistant")[1])
|
85 |
+
```
|
|