File size: 2,421 Bytes
af71584 4c0932e 6e1e776 ec9d819 cce0122 9bbdc3b b425eea cce0122 9bbdc3b b425eea 6e1e776 947480c af71584 f7c14e4 af71584 e067c58 af71584 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 |
import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed, pipeline
title = "SantaCoder 🎅 bash/shell 🐚 Completion"
description = "This is a subspace to make code generation with [SantaCoder fine-tuned on The Stack bash/shell](https://huggingface.co/mrm8488/santacoder-finetuned-the-stack-bash-4)"
EXAMPLE_0 = "#!/bin/bash\n# This script removes files larger than 2MB in the current folder\nfind ."
EXAMPLE_1 = "#!/bin/bash\n\n# This script send an email\nto=”admin@example.com”\nsubject=”Greeting”\nmsg=”Welcome to our site”\n"
EXAMPLE_3 = "#!/bin/bash\n# This script convert avi files to mp4\nfor filename in $(ls *.avi); do\n"
EXAMPLE_4 = "#!/bin/bash\nsource=$1\ndest=$2\n# copy source on dest\n"
EXAMPLE_5 = """#!/bin/bash\n\n# This script check if the arg passed as first arg is a founder of huggingface\nfounders_array=("julien" "thom" "clem")"""
examples = [[EXAMPLE_0, 14, 0.6, 42], [EXAMPLE_1, 28, 0.6, 42], [EXAMPLE_3, 46, 0.6, 42], [EXAMPLE_4, 35, 0.6, 43], [EXAMPLE_5, 70, 0.6, 43]]
tokenizer = AutoTokenizer.from_pretrained("mrm8488/santacoder-finetuned-the-stack-bash-4")
model = AutoModelForCausalLM.from_pretrained("mrm8488/santacoder-finetuned-the-stack-bash-4", trust_remote_code=True)
def code_generation(gen_prompt, max_tokens, temperature=0.6, seed=42):
set_seed(seed)
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
generated_text = pipe(gen_prompt, do_sample=True, top_p=0.95, temperature=temperature, max_new_tokens=max_tokens)[0]['generated_text']
return generated_text
iface = gr.Interface(
fn=code_generation,
inputs=[
gr.Textbox(lines=10, label="Input code"),
gr.inputs.Slider(
minimum=8,
maximum=256,
step=1,
default=8,
label="Number of tokens to generate",
),
gr.inputs.Slider(
minimum=0,
maximum=2,
step=0.1,
default=0.6,
label="Temperature",
),
gr.inputs.Slider(
minimum=0,
maximum=1000,
step=1,
default=42,
label="Random seed to use for the generation"
)
],
outputs=gr.Textbox(label="Predicted code", lines=10),
examples=examples,
layout="horizontal",
theme="peach",
description=description,
title=title
)
iface.launch() |