Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,7 +1,29 @@
|
|
1 |
import gradio as gr
|
2 |
|
3 |
-
|
4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
|
6 |
demo = gr.Interface(fn=greet, inputs="text", outputs="text")
|
7 |
demo.launch()
|
|
|
1 |
import gradio as gr
|
2 |
|
3 |
+
from unsloth import FastLanguageModel
|
4 |
+
model, tokenizer = FastLanguageModel.from_pretrained(
|
5 |
+
model_name = "benjamintli/llama3.2_abc_fine_tune", # YOUR MODEL YOU USED FOR TRAINING
|
6 |
+
max_seq_length = 2048,
|
7 |
+
dtype = None,
|
8 |
+
load_in_4bit = True,
|
9 |
+
)
|
10 |
+
|
11 |
+
def greet(input):
|
12 |
+
messages = [
|
13 |
+
{"role": "system", "content": "you are a helpful assistant that generates ABC notation music"},
|
14 |
+
{"role": "user", "content": f"{input}"},
|
15 |
+
]
|
16 |
+
inputs = tokenizer.apply_chat_template(
|
17 |
+
messages,
|
18 |
+
tokenize = True,
|
19 |
+
add_generation_prompt = True, # Must add for generation
|
20 |
+
return_tensors = "pt",
|
21 |
+
)
|
22 |
+
|
23 |
+
outputs = model.generate(input_ids = inputs, max_new_tokens = 128, use_cache = True,
|
24 |
+
temperature = 1.5, min_p = 0.1)
|
25 |
+
response = tokenizer.batch_decode(outputs)
|
26 |
+
return response
|
27 |
|
28 |
demo = gr.Interface(fn=greet, inputs="text", outputs="text")
|
29 |
demo.launch()
|