Model Card for Model ID

Model Details

Model Description

Generating Monthly Weight Increase Rate Plan~

  • Developed by: [More Information Needed]
  • Model type: adapter
  • Language(s) (NLP): korean
  • Finetuned from model: Polyglot-ko-1.3b

Usage Example

import time
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

# ๐Ÿ”ง Model + checkpoint config
base_model_name = "EleutherAI/polyglot-ko-1.3b"
lora_model_path = "jwywoo/burnfit-lora-v2"

# โœ… Load tokenizer
tokenizer = AutoTokenizer.from_pretrained(lora_model_path)
tokenizer.add_special_tokens({"additional_special_tokens": ["<END>"]})
tokenizer.eos_token = "<END>"

# โœ… Load base model + LoRA adapter
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
base_model = AutoModelForCausalLM.from_pretrained(base_model_name).to(device)
base_model.resize_token_embeddings(len(tokenizer))
model = PeftModel.from_pretrained(base_model, lora_model_path).to(device)
model.eval()


# For Newbies if you are using colab put this part in different cell
# ๐Ÿ”  Prompt example
prompt = """## ์งˆ๋ฌธ
- 5/3/1 ํ”„๋กœ๊ทธ๋žจ
- **์šด๋™๊ฒฝํ—˜**: 2
- **์„ฑ๋ณ„**: ์—ฌ์„ฑ
- **์šด๋™๋ชฉํ‘œ**: 3
- **1RM**:
  - ๋ฒค์น˜ํ”„๋ ˆ์Šค: 60
  - ์Šค์ฟผํŠธ: 70
  - ๋ฐ๋“œ๋ฆฌํ”„ํŠธ: 75
  - ์˜ค๋ฒ„ํ—ค๋“œํ”„๋ ˆ์Šค: 40

## ๋‹ต๋ณ€
"""

# โฑ๏ธ Measure time
start = time.time()

inputs = tokenizer(prompt, return_tensors="pt").to(device)
inputs.pop("token_type_ids", None)

eos_token_id = tokenizer.convert_tokens_to_ids("<END>")
outputs = model.generate(
    **inputs,
    max_new_tokens=300,
    eos_token_id=eos_token_id,
    do_sample=False,
    temperature=0.7,
    top_p=0.9,
    repetition_penalty=1.1
)

# ๐Ÿ“„ Decode and trim
generated = tokenizer.decode(outputs[0])
generated = generated.split("<END>")[0].strip() + "\n<END>"

elapsed = time.time() - start
print(f"\nโฑ๏ธ Inference Time: {elapsed:.2f} seconds")
print("\n๐Ÿง  Generated Output:\n")
print(generated)

Example Input

## ์งˆ๋ฌธ
- 5/3/1ํ”„๋กœ๊ทธ๋žจ
- **์šด๋™๊ฒฝํ—˜**: 3
- **์„ฑ๋ณ„**: ๋‚จ์„ฑ
- **์šด๋™๋ชฉํ‘œ**: 2
- **1RM**:
  - ๋ฒค์น˜ํ”„๋ ˆ์Šค: 60kg
  - ์Šค์ฟผํŠธ: 70kg
  - ๋ฐ๋“œ๋ฆฌํ”„ํŠธ: 70kg
  - ์˜ค๋ฒ„ํ—ค๋“œํ”„๋ ˆ์Šค: 70kg

## ๋‹ต๋ณ€

Example Output

## ์งˆ๋ฌธ
- 5/3/1ํ”„๋กœ๊ทธ๋žจ
- **์šด๋™๊ฒฝํ—˜**: 3
- **์„ฑ๋ณ„**: ๋‚จ์„ฑ
- **์šด๋™๋ชฉํ‘œ**: 2
- **1RM**:
  - ๋ฒค์น˜ํ”„๋ ˆ์Šค: 60kg
  - ์Šค์ฟผํŠธ: 70kg
  - ๋ฐ๋“œ๋ฆฌํ”„ํŠธ: 70kg
  - ์˜ค๋ฒ„ํ—ค๋“œํ”„๋ ˆ์Šค: 70kg

## ๋‹ต๋ณ€
{
  "program": "5/3/1ํ”„๋กœ๊ทธ๋žจ",
  "init_weight_rate": "55",
  "increase_rate_week": "10",
  "increase_rate_set": "10",
  "deloading_rate": "30",
  "weekly_weight_increase_plan": "Week1\nSET1 55%x5\nSET2 65%x5\nSET3 75%x5+\n\nWeek2\nSET1 65%x3\nSET2 75%x3\nSET3 85%x3+\n\nWeek3\nSET1 75%x5\nSET2 80%x3\nSET3 85%x1+\n\nWeek4\nSET1 30%x5\nSET2 40%x5\nSET3 50%x5\n"
}
  • PEFT 0.15.2
Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for jwywoo/burnfit-lora-v2

Adapter
(28)
this model