Spaces:
Running
Running
File size: 4,290 Bytes
df96e38 |
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 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 |
import sys
import fire
import torch
# from peft import PeftModel
import transformers
import gradio as gr
assert (
"LlamaTokenizer" in transformers._import_structure["models.llama"]
), "LLaMA is now in HuggingFace's main branch.\nPlease reinstall it: pip uninstall transformers && pip install git+https://github.com/huggingface/transformers.git"
from transformers import LlamaTokenizer, LlamaForCausalLM, GenerationConfig
if torch.cuda.is_available():
device = "cuda"
else:
device = "cpu"
try:
if torch.backends.mps.is_available():
device = "mps"
except:
pass
def main(
load_8bit: bool = False,
base_model: str = "/path/to/llama-7B/hf/ft/checkpoint-300",
# lora_weights: str = "tloen/alpaca-lora-7b",
):
assert base_model, (
"Please specify a --base_model, e.g. --base_model='decapoda-research/llama-7b-hf'"
)
tokenizer = LlamaTokenizer.from_pretrained(base_model)
if device == "cuda":
model = LlamaForCausalLM.from_pretrained(
base_model,
load_in_8bit=load_8bit,
torch_dtype=torch.float16,
device_map="auto",
)
elif device == "mps":
model = LlamaForCausalLM.from_pretrained(
base_model,
device_map={"": device},
torch_dtype=torch.float16,
)
# unwind broken decapoda-research config
model.config.pad_token_id = tokenizer.pad_token_id = 0 # unk
model.config.bos_token_id = 1
model.config.eos_token_id = 2
if not load_8bit:
model.half() # seems to fix bugs for some users.
model.eval()
if torch.__version__ >= "2" and sys.platform != "win32":
model = torch.compile(model)
def evaluate(
instruction,
input=None,
temperature=0.6,
top_p=0.9,
top_k=40,
num_beams=4,
max_new_tokens=512,
**kwargs,
):
prompt = generate_prompt(instruction, input)
inputs = tokenizer(prompt, return_tensors="pt")
input_ids = inputs["input_ids"].to(device)
generation_config = GenerationConfig(
temperature=temperature,
top_p=top_p,
top_k=top_k,
num_beams=num_beams,
**kwargs,
)
with torch.no_grad():
generation_output = model.generate(
input_ids=input_ids,
generation_config=generation_config,
return_dict_in_generate=True,
output_scores=True,
max_new_tokens=max_new_tokens,
)
s = generation_output.sequences[0]
output = tokenizer.decode(s)
return output.split("### Response:")[1].strip()
gr.Interface(
fn=evaluate,
inputs=[
gr.components.Textbox(
lines=2, label="Instruction", placeholder="Tell me about alpacas."
),
gr.components.Textbox(lines=2, label="Input", placeholder="none"),
gr.components.Slider(minimum=0, maximum=1, value=0.1, label="Temperature"),
gr.components.Slider(minimum=0, maximum=1, value=0.75, label="Top p"),
gr.components.Slider(
minimum=0, maximum=100, step=1, value=40, label="Top k"
),
gr.components.Slider(minimum=1, maximum=4, step=1, value=4, label="Beams"),
gr.components.Slider(
minimum=1, maximum=2000, step=1, value=128, label="Max tokens"
),
],
outputs=[
gr.inputs.Textbox(
lines=5,
label="Output",
)
],
title="Llama-X",
description="Improve LLaMA model to follow instructions.",
).launch(share=True)
def generate_prompt(instruction, input=None):
if input:
return f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
### Instruction:
{instruction}
### Input:
{input}
### Response:
"""
else:
return f"""Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{instruction}
### Response:
"""
if __name__ == "__main__":
fire.Fire(main) |