Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -37,18 +37,15 @@ def format_prompt_grammar(message, history):
|
|
37 |
|
38 |
|
39 |
|
40 |
-
def generate(prompt, history, system_prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0
|
41 |
temperature = float(temperature)
|
42 |
if temperature < 1e-2: temperature = 1e-2
|
43 |
top_p = float(top_p)
|
44 |
|
45 |
generate_kwargs = dict(temperature=temperature, max_new_tokens=max_new_tokens, top_p=top_p, repetition_penalty=repetition_penalty, do_sample=True, seed=42,)
|
46 |
|
47 |
-
|
48 |
-
|
49 |
-
formatted_prompt = format_prompt_grammar(f"Corrected Sentence: {prompt}", history)
|
50 |
-
else:
|
51 |
-
formatted_prompt = format_prompt(f"{system_prompt} {prompt}", history)
|
52 |
print("\nPROMPT: \n\t" + formatted_prompt)
|
53 |
|
54 |
# Generate text from the HF inference
|
@@ -67,8 +64,7 @@ additional_inputs=[
|
|
67 |
gr.Slider( label="Temperature", value=0.9, minimum=0.0, maximum=1.0, step=0.05, interactive=True, info="Higher values produce more diverse outputs", ),
|
68 |
gr.Slider( label="Max new tokens", value=256, minimum=0, maximum=1048, step=64, interactive=True, info="The maximum numbers of new tokens", ),
|
69 |
gr.Slider( label="Top-p (nucleus sampling)", value=0.90, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more low-probability tokens", ),
|
70 |
-
gr.Slider( label="Repetition penalty", value=1.2, minimum=1.0, maximum=2.0, step=0.05, interactive=True, info="Penalize repeated tokens", )
|
71 |
-
gr.Checkbox( label="GEC", value="GEC Mode?", interactive=True, )
|
72 |
]
|
73 |
|
74 |
examples=['Give me the grammatically correct version of the sentence: "We shood buy an car"', "Give me an example exam question testing students on square roots on basic integers", "Would this block of HTML code run?\n```\n\n```"]
|
|
|
37 |
|
38 |
|
39 |
|
40 |
+
def generate(prompt, history, system_prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
|
41 |
temperature = float(temperature)
|
42 |
if temperature < 1e-2: temperature = 1e-2
|
43 |
top_p = float(top_p)
|
44 |
|
45 |
generate_kwargs = dict(temperature=temperature, max_new_tokens=max_new_tokens, top_p=top_p, repetition_penalty=repetition_penalty, do_sample=True, seed=42,)
|
46 |
|
47 |
+
#formatted_prompt = format_prompt_grammar(f"Corrected Sentence: {prompt}", history)
|
48 |
+
formatted_prompt = format_prompt(f"{system_prompt} {prompt}", history)
|
|
|
|
|
|
|
49 |
print("\nPROMPT: \n\t" + formatted_prompt)
|
50 |
|
51 |
# Generate text from the HF inference
|
|
|
64 |
gr.Slider( label="Temperature", value=0.9, minimum=0.0, maximum=1.0, step=0.05, interactive=True, info="Higher values produce more diverse outputs", ),
|
65 |
gr.Slider( label="Max new tokens", value=256, minimum=0, maximum=1048, step=64, interactive=True, info="The maximum numbers of new tokens", ),
|
66 |
gr.Slider( label="Top-p (nucleus sampling)", value=0.90, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more low-probability tokens", ),
|
67 |
+
gr.Slider( label="Repetition penalty", value=1.2, minimum=1.0, maximum=2.0, step=0.05, interactive=True, info="Penalize repeated tokens", )
|
|
|
68 |
]
|
69 |
|
70 |
examples=['Give me the grammatically correct version of the sentence: "We shood buy an car"', "Give me an example exam question testing students on square roots on basic integers", "Would this block of HTML code run?\n```\n\n```"]
|