|
|
--- |
|
|
base_model: SaintHoney/PersonalManV1.0 |
|
|
tags: |
|
|
- text-generation-inference |
|
|
- transformers |
|
|
- unsloth |
|
|
- qwen2 |
|
|
- trl |
|
|
- sft |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
datasets: |
|
|
- diabolic6045/open-ocra-alpaca-cleaned |
|
|
- HashTag766/SMART-Goals-Validation |
|
|
--- |
|
|
|
|
|
# Overview |
|
|
#### Finetuned Qwen2.5-3B |
|
|
#### the training was for increasing the model capabilities on Instruction following and specific data. |
|
|
#### Training Time : 14.5h |
|
|
|
|
|
### Datasets |
|
|
#### SMART-Goals-Validation------[https://huggingface.co/datasets/HashTag766/SMART-Goals-Validation] |
|
|
#### open-ocra-alpaca-cleaned----[https://huggingface.co/datasets/diabolic6045/open-ocra-alpaca-cleaned] only on 120000k examples |
|
|
|
|
|
|
|
|
# Uploaded model |
|
|
|
|
|
- **Developed by:** HashTag766 |
|
|
- **License:** apache-2.0 |
|
|
- **Finetuned from model :** SaintHoney/PersonalManV1.0 |
|
|
|
|
|
## The code used for finetuning |
|
|
|
|
|
```python |
|
|
%%capture |
|
|
!pip install pip3-autoremove |
|
|
!pip-autoremove torch torchvision torchaudio -y |
|
|
!pip install torch torchvision torchaudio xformers --index-url https://download.pytorch.org/whl/cu121 |
|
|
!pip install unsloth |
|
|
|
|
|
--------------------------------------------------------------------------------------------- |
|
|
|
|
|
from kaggle_secrets import UserSecretsClient |
|
|
user_secrets = UserSecretsClient() # from kaggle_secrets import UserSecretsClient |
|
|
hugging_face_token = user_secrets.get_secret("HF-Token") |
|
|
|
|
|
# Login to Hugging Face |
|
|
from huggingface_hub import login # Lets you login to API |
|
|
login(hugging_face_token) # from huggingface_hub import login |
|
|
|
|
|
--------------------------------------------------------------------------------------------- |
|
|
|
|
|
from unsloth import FastLanguageModel |
|
|
import torch |
|
|
max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally! |
|
|
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+ |
|
|
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False. |
|
|
|
|
|
model, tokenizer = FastLanguageModel.from_pretrained( |
|
|
model_name = "SaintHoney/PersonalManV1.0", |
|
|
max_seq_length = max_seq_length, |
|
|
dtype = dtype, |
|
|
load_in_4bit = load_in_4bit, |
|
|
) |
|
|
|
|
|
--------------------------------------------------------------------------------------------- |
|
|
|
|
|
model = FastLanguageModel.get_peft_model( |
|
|
model, |
|
|
r = 16, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128 |
|
|
target_modules = ["q_proj", "k_proj", "v_proj", "o_proj", |
|
|
"gate_proj", "up_proj", "down_proj",], |
|
|
lora_alpha = 16, |
|
|
lora_dropout = 0, # Supports any, but = 0 is optimized |
|
|
bias = "none", # Supports any, but = "none" is optimized |
|
|
# [NEW] "unsloth" uses 30% less VRAM, fits 2x larger batch sizes! |
|
|
use_gradient_checkpointing = "unsloth", # True or "unsloth" for very long context |
|
|
random_state = 3407, |
|
|
use_rslora = False, # We support rank stabilized LoRA |
|
|
loftq_config = None, # And LoftQ |
|
|
) |
|
|
|
|
|
--------------------------------------------------------------------------------------------- |
|
|
|
|
|
alpaca_prompt = """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: |
|
|
{} |
|
|
|
|
|
### Input: |
|
|
{} |
|
|
|
|
|
### Response: |
|
|
{}""" |
|
|
|
|
|
EOS_TOKEN = tokenizer.eos_token # Must add EOS_TOKEN |
|
|
def formatting_prompts_func(examples): |
|
|
instructions = examples["instruction"] |
|
|
inputs = examples["input"] |
|
|
outputs = examples["output"] |
|
|
texts = [] |
|
|
for instruction, input, output in zip(instructions, inputs, outputs): |
|
|
# Must add EOS_TOKEN, otherwise your generation will go on forever! |
|
|
text = alpaca_prompt.format(instruction, input, output) + EOS_TOKEN |
|
|
texts.append(text) |
|
|
return { "text" : texts, } |
|
|
pass |
|
|
|
|
|
from datasets import load_dataset |
|
|
dataset = load_dataset("HashTag766/SMART-Goals-Validation", split = "train") # specify here the number of examples from dataset |
|
|
dataset = dataset.map(formatting_prompts_func, batched = True,) |
|
|
|
|
|
--------------------------------------------------------------------------------------------- |
|
|
|
|
|
from trl import SFTTrainer |
|
|
from transformers import TrainingArguments, DataCollatorForSeq2Seq |
|
|
from unsloth import is_bfloat16_supported |
|
|
|
|
|
trainer = SFTTrainer( |
|
|
model = model, |
|
|
tokenizer = tokenizer, |
|
|
train_dataset = dataset, |
|
|
dataset_text_field = "text", |
|
|
max_seq_length = max_seq_length, |
|
|
data_collator = DataCollatorForSeq2Seq(tokenizer = tokenizer), |
|
|
dataset_num_proc = 2, |
|
|
packing = False, # Can make training 5x faster for short sequences. |
|
|
args = TrainingArguments( |
|
|
per_device_train_batch_size = 2, |
|
|
gradient_accumulation_steps = 4, |
|
|
warmup_steps = 5, |
|
|
num_train_epochs = 3, # Set this for 1 full training run. |
|
|
# max_steps = 60, |
|
|
learning_rate = 2e-4, |
|
|
fp16 = not is_bfloat16_supported(), |
|
|
bf16 = is_bfloat16_supported(), |
|
|
logging_steps = 1, |
|
|
optim = "adamw_8bit", |
|
|
weight_decay = 0.01, |
|
|
lr_scheduler_type = "linear", |
|
|
seed = 3407, |
|
|
output_dir = "outputs", |
|
|
report_to = "none", # Use this for WandB etc |
|
|
), |
|
|
) |
|
|
|
|
|
trainer_stats = trainer.train() |
|
|
--------------------------------------------------------------------------------------------- |
|
|
|
|
|
model.push_to_hub("hf/model...", token = "...") # Online saving |
|
|
tokenizer.push_to_hub("hf/model...", token = "...") # Online saving |
|
|
|
|
|
``` |
|
|
|
|
|
This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. |
|
|
|
|
|
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth) |
