# -*- coding: utf-8 -*-
"""Copy of Alpaca + Llama-3 8b full example.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/12GTGPtaZvutZlE2GUHmeXVrvq2dgJdu6

## To run this, press "*Runtime*" and press "*Run all*" on a **free** Tesla T4 Google Colab instance!

To install Unsloth on your own computer, follow the installation instructions on our Github page [here](https://github.com/unslothai/unsloth#installation-instructions---conda).

You will learn how to do [data prep](#Data), how to [train](#Train), how to [run the model](#Inference), & [how to save it](#Save) (eg for Llama.cpp).

**Llama-3 8b is trained on a crazy 15 trillion tokens! Llama-2 was 2 trillion.**
"""

# Commented out IPython magic to ensure Python compatibility.
# %%capture
# # Installs Unsloth, Xformers (Flash Attention) and all other packages!
# !pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
# !pip install --no-deps "xformers<0.0.26" trl peft accelerate bitsandbytes

token = "" #HF Token

"""* We support Llama, Mistral, CodeLlama, TinyLlama, Vicuna, Open Hermes etc
* And Yi, Qwen ([llamafied](https://huggingface.co./models?sort=trending&search=qwen+llama)), Deepseek, all Llama, Mistral derived archs.
* We support 16bit LoRA or 4bit QLoRA. Both 2x faster.
* `max_seq_length` can be set to anything, since we do automatic RoPE Scaling via [kaiokendev's](https://kaiokendev.github.io/til) method.
* [**NEW**] With [PR 26037](https://github.com/huggingface/transformers/pull/26037), we support downloading 4bit models **4x faster**! [Our repo](https://huggingface.co./unsloth) has Llama, Mistral 4bit models.
"""

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.

# 4bit pre quantized models we support for 4x faster downloading + no OOMs.
fourbit_models = [
    "unsloth/mistral-7b-bnb-4bit",
    "unsloth/mistral-7b-instruct-v0.2-bnb-4bit",
    "unsloth/llama-2-7b-bnb-4bit",
    "unsloth/gemma-7b-bnb-4bit",
    "unsloth/gemma-7b-it-bnb-4bit", # Instruct version of Gemma 7b
    "unsloth/gemma-2b-bnb-4bit",
    "unsloth/gemma-2b-it-bnb-4bit", # Instruct version of Gemma 2b
    "unsloth/llama-3-8b-bnb-4bit", # [NEW] 15 Trillion token Llama-3
] # More models at https://huggingface.co./unsloth

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "unsloth/llama-3-8b-bnb-4bit",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
    # token = token, # use one if using gated models like meta-llama/Llama-2-7b-hf
)

"""We now add LoRA adapters so we only need to update 1 to 10% of all parameters!"""

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
)

"""<a name="Data"></a>
### Data Prep
We now use the Alpaca dataset from [yahma](https://huggingface.co./datasets/yahma/alpaca-cleaned), which is a filtered version of 52K of the original [Alpaca dataset](https://crfm.stanford.edu/2023/03/13/alpaca.html). You can replace this code section with your own data prep.

**[NOTE]** To train only on completions (ignoring the user's input) read TRL's docs [here](https://huggingface.co./docs/trl/sft_trainer#train-on-completions-only).

**[NOTE]** Remember to add the **EOS_TOKEN** to the tokenized output!! Otherwise you'll get infinite generations!

If you want to use the `ChatML` template for ShareGPT datasets, try our conversational [notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing).

For text completions like novel writing, try this [notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing).
"""

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("yahma/alpaca-cleaned", split = "train")
dataset = dataset.map(formatting_prompts_func, batched = True,)

"""<a name="Train"></a>
### Train the model
Now let's use Huggingface TRL's `SFTTrainer`! More docs here: [TRL SFT docs](https://huggingface.co./docs/trl/sft_trainer). We do 60 steps to speed things up, but you can set `num_train_epochs=1` for a full run, and turn off `max_steps=None`. We also support TRL's `DPOTrainer`!
"""

from trl import SFTTrainer
from transformers import TrainingArguments

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 torch.cuda.is_bf16_supported(),
        bf16 = torch.cuda.is_bf16_supported(),
        logging_steps = 1,
        optim = "adamw_8bit",
        weight_decay = 0.01,
        lr_scheduler_type = "linear",
        seed = 3407,
        output_dir = "outputs",
    ),
)

#@title Show current memory stats
gpu_stats = torch.cuda.get_device_properties(0)
start_gpu_memory = round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3)
max_memory = round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3)
print(f"GPU = {gpu_stats.name}. Max memory = {max_memory} GB.")
print(f"{start_gpu_memory} GB of memory reserved.")

trainer_stats = trainer.train()

#@title Show final memory and time stats
used_memory = round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3)
used_memory_for_lora = round(used_memory - start_gpu_memory, 3)
used_percentage = round(used_memory         /max_memory*100, 3)
lora_percentage = round(used_memory_for_lora/max_memory*100, 3)
print(f"{trainer_stats.metrics['train_runtime']} seconds used for training.")
print(f"{round(trainer_stats.metrics['train_runtime']/60, 2)} minutes used for training.")
print(f"Peak reserved memory = {used_memory} GB.")
print(f"Peak reserved memory for training = {used_memory_for_lora} GB.")
print(f"Peak reserved memory % of max memory = {used_percentage} %.")
print(f"Peak reserved memory for training % of max memory = {lora_percentage} %.")

"""<a name="Inference"></a>
### Inference
Let's run the model! You can change the instruction and input - leave the output blank!
"""

# alpaca_prompt = Copied from above
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
inputs = tokenizer(
[
    alpaca_prompt.format(
        "Continue the fibonnaci sequence.", # instruction
        "1, 1, 2, 3, 5, 8", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
tokenizer.batch_decode(outputs)

""" You can also use a `TextStreamer` for continuous inference - so you can see the generation token by token, instead of waiting the whole time!"""

# alpaca_prompt = Copied from above
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
inputs = tokenizer(
[
    alpaca_prompt.format(
        "Continue the fibonnaci sequence.", # instruction
        "1, 1, 2, 3, 5, 8", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 128)

"""<a name="Save"></a>
### Saving, loading finetuned models
To save the final model as LoRA adapters, either use Huggingface's `push_to_hub` for an online save or `save_pretrained` for a local save.

**[NOTE]** This ONLY saves the LoRA adapters, and not the full model. To save to 16bit or GGUF, scroll down!
"""

#model.save_pretrained("lora_model") # Local saving
#tokenizer.save_pretrained("lora_model")
model.push_to_hub("ArunKr/LLama3-LoRA", token = token) # Online saving
tokenizer.push_to_hub("ArunKr/LLama3-LoRA", token = token) # Online saving

"""Now if you want to load the LoRA adapters we just saved for inference, set `False` to `True`:"""

if False:
    from unsloth import FastLanguageModel
    model, tokenizer = FastLanguageModel.from_pretrained(
        model_name = "lora_model", # YOUR MODEL YOU USED FOR TRAINING
        max_seq_length = max_seq_length,
        dtype = dtype,
        load_in_4bit = load_in_4bit,
    )
    FastLanguageModel.for_inference(model) # Enable native 2x faster inference

# alpaca_prompt = You MUST copy from above!

inputs = tokenizer(
[
    alpaca_prompt.format(
        "What is a famous tall tower in Paris?", # instruction
        "", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 64, use_cache = True)
tokenizer.batch_decode(outputs)

"""You can also use Hugging Face's `AutoModelForPeftCausalLM`. Only use this if you do not have `unsloth` installed. It can be hopelessly slow, since `4bit` model downloading is not supported, and Unsloth's **inference is 2x faster**."""

if False:
    # I highly do NOT suggest - use Unsloth if possible
    from peft import AutoPeftModelForCausalLM
    from transformers import AutoTokenizer
    model = AutoPeftModelForCausalLM.from_pretrained(
        "lora_model", # YOUR MODEL YOU USED FOR TRAINING
        load_in_4bit = load_in_4bit,
    )
    tokenizer = AutoTokenizer.from_pretrained("lora_model")

"""### Saving to float16 for VLLM

We also support saving to `float16` directly. Select `merged_16bit` for float16 or `merged_4bit` for int4. We also allow `lora` adapters as a fallback. Use `push_to_hub_merged` to upload to your Hugging Face account! You can go to https://huggingface.co./settings/tokens for your personal tokens.
"""

# Merge to 16bit
if False: model.save_pretrained_merged("model", tokenizer, save_method = "merged_16bit",)
if True: model.push_to_hub_merged("ArunKr/LLama3-LoRA", tokenizer, save_method = "merged_16bit", token = token)

# Merge to 4bit
if False: model.save_pretrained_merged("model", tokenizer, save_method = "merged_4bit",)
if True: model.push_to_hub_merged("ArunKr/LLama3-LoRA", tokenizer, save_method = "merged_4bit_forced", token = token)

# Just LoRA adapters
if False: model.save_pretrained_merged("model", tokenizer, save_method = "lora",)
if True: model.push_to_hub_merged("ArunKr/LLama3-LoRA", tokenizer, save_method = "lora", token = token)

"""### GGUF / llama.cpp Conversion
To save to `GGUF` / `llama.cpp`, we support it natively now! We clone `llama.cpp` and we default save it to `q8_0`. We allow all methods like `q4_k_m`. Use `save_pretrained_gguf` for local saving and `push_to_hub_gguf` for uploading to HF.

Some supported quant methods (full list on our [Wiki page](https://github.com/unslothai/unsloth/wiki#gguf-quantization-options)):
* `q8_0` - Fast conversion. High resource use, but generally acceptable.
* `q4_k_m` - Recommended. Uses Q6_K for half of the attention.wv and feed_forward.w2 tensors, else Q4_K.
* `q5_k_m` - Recommended. Uses Q6_K for half of the attention.wv and feed_forward.w2 tensors, else Q5_K.
"""

# Save to 8bit Q8_0
if False: model.save_pretrained_gguf("model", tokenizer,)
if True: model.push_to_hub_gguf("ArunKr/LLama3-LoRA", tokenizer, token = token)

# Save to 16bit GGUF
if False: model.save_pretrained_gguf("model", tokenizer, quantization_method = "f16")
if True: model.push_to_hub_gguf("ArunKr/LLama3-LoRA", tokenizer, quantization_method = "f16", token = token)

# Save to q4_k_m GGUF
if False: model.save_pretrained_gguf("model", tokenizer, quantization_method = "q4_k_m")
if True: model.push_to_hub_gguf("ArunKr/LLama3-LoRA", tokenizer, quantization_method = "q4_k_m", token = token)

"""Now, use the `model-unsloth.gguf` file or `model-unsloth-Q4_K_M.gguf` file in `llama.cpp` or a UI based system like `GPT4All`. You can install GPT4All by going [here](https://gpt4all.io/index.html).

And we're done! If you have any questions on Unsloth, we have a [Discord](https://discord.gg/u54VK8m8tk) channel! If you find any bugs or want to keep updated with the latest LLM stuff, or need help, join projects etc, feel free to join our Discord!

Some other links:
1. Zephyr DPO 2x faster [free Colab](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing)
2. Llama 7b 2x faster [free Colab](https://colab.research.google.com/drive/1lBzz5KeZJKXjvivbYvmGarix9Ao6Wxe5?usp=sharing)
3. TinyLlama 4x faster full Alpaca 52K in 1 hour [free Colab](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing)
4. CodeLlama 34b 2x faster [A100 on Colab](https://colab.research.google.com/drive/1y7A0AxE3y8gdj4AVkl2aZX47Xu3P1wJT?usp=sharing)
5. Mistral 7b [free Kaggle version](https://www.kaggle.com/code/danielhanchen/kaggle-mistral-7b-unsloth-notebook)
6. We also did a [blog](https://huggingface.co./blog/unsloth-trl) with 🤗 HuggingFace, and we're in the TRL [docs](https://huggingface.co./docs/trl/main/en/sft_trainer#accelerate-fine-tuning-2x-using-unsloth)!
7. `ChatML` for ShareGPT datasets, [conversational notebook](https://colab.research.google.com/drive/1Aau3lgPzeZKQ-98h69CCu1UJcvIBLmy2?usp=sharing)
8. Text completions like novel writing [notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing)
"""