10G显存，使用Unsloth微调Qwen2并使用Ollama推理

!pip install "unsloth[cu121-torch230] @ git+https://github.com/unslothai/unsloth.git"

from unsloth import FastLanguageModel
import torch
from modelscope import snapshot_download
max_seq_length = 2048 
dtype = torch.bfloat16
load_in_4bit = True 
model_dir=snapshot_download("qwen/Qwen2-7B")
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = model_dir, 
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
)

max_seq_length = 2048

dtype = torch.bfloat16

load_in_4bit = True

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
)

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

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

from unsloth.chat_templates import get_chat_template
tokenizer = get_chat_template(
    tokenizer,
    chat_template = "chatml", # Supports zephyr, chatml, mistral, llama, alpaca, vicuna, vicuna_old, unsloth
    mapping = {"role" : "from", "content" : "value", "user" : "human", "assistant" : "gpt"}, # ShareGPT style
)
def formatting_prompts_func(examples):
    convos = examples["conversations"]
    texts = [tokenizer.apply_chat_template(convo, tokenize = False, add_generation_prompt = False) for convo in convos]
    return { "text" : texts, }
pass
from modelscope.msdatasets import MsDataset
dataset =  MsDataset.load('OmniData/guanaco-sharegpt-style', split = "train")
dataset = dataset.map(formatting_prompts_func, batched = True,)

from trl import SFTTrainer
from transformers import TrainingArguments
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,
    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,
        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",
    ),
)

per_device_train_batch_size = 2,

gradient_accumulation_steps = 4,

max_steps = 60, # num_train_epochs = 1,

learning_rate = 2e-4,

trainer_stats = trainer.train()

from unsloth.chat_templates import get_chat_template
tokenizer = get_chat_template(
    tokenizer,
    chat_template = "chatml", # Supports zephyr, chatml, mistral, llama, alpaca, vicuna, vicuna_old, unsloth
    mapping = {"role" : "from", "content" : "value", "user" : "human", "assistant" : "gpt"}, # ShareGPT style
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
messages = [
    {"from": "human", "value": "杭州的省会在哪里？"},
]
inputs = tokenizer.apply_chat_template(
    messages,
    tokenize = True,
    add_generation_prompt = True, # Must add for generation
    return_tensors = "pt",
).to("cuda")
outputs = model.generate(input_ids = inputs, max_new_tokens = 64, use_cache = True)
tokenizer.batch_decode(outputs)

FastLanguageModel.for_inference(model) # Enable native 2x faster inference
messages = [
    {"from": "human", "value": "杭州的省会在哪里？"},
]
inputs = tokenizer.apply_chat_template(
    messages,
    tokenize = True,
    add_generation_prompt = True, # Must add for generation
    return_tensors = "pt",
).to("cuda")
from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer, skip_prompt = True)
_ = model.generate(input_ids = inputs, streamer = text_streamer, max_new_tokens = 128, use_cache = True)

model.save_pretrained("lora_model") # Local saving
tokenizer.save_pretrained("lora_model")

if True: model.save_pretrained_gguf("model", tokenizer, quantization_method = "q4_k_m")

print(tokenizer._ollama_modelfile)

git clone https://www.modelscope.cn/modelscope/ollama-linux.git
cd ollama-linux
sudo chmod 777 ./ollama-modelscope-install.sh
./ollama-modelscope-install.sh

ollama serve

!ollama create unsloth_qwen2 -f /mnt/workspace/model/Modelfile

ollama run unsloth_qwen2