---
base_model: unsloth/qwen2-0.5b-bnb-4bit
language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
- sft
---

## From the 10 refinement started answering the last fibonassi numbers wrong again, and now in this 13 refinement using Lora is back to answering correctly and writing the code to train an offline AI correctly.

1 - Question

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(
        "I need train a AI offline on my computer, give me a code good for this case.", # instruction
        "", # 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 = 4096)

2 - Answer

Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Input:
I need train a AI offline on my computer, give me a code good for this case.

### Output:
Sure, here's a simple example of how you can train an AI model on a computer using Python and TensorFlow. This is a very basic example and doesn't include many features that are commonly used in real-world applications. 

Firstly, you need to install the necessary libraries. You can do this using pip:

```python
pip install tensorflow
```

Then, you can start training your model:

```python
import tensorflow as tf
from tensorflow.keras import models, layers

# define the model
model = tf.keras.models.Sequential([
    tf.keras.layers.Dense(64, activation='relu', input_shape=(100,)),
    tf.keras.layers.Dense(64, activation='relu'),
    tf.keras.layers.Dense(10)
])

# compile the model
model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

# train the model
model.fit(X_train, y_train, epochs=10)
```

Here, `X_train` and `y_train` are your training data, and `X_train.shape` and `y_train.shape` are the shape of your training data.

Please note that this is a very simple example and you may need to adjust the model architecture, number of layers, number of neurons in each layer, and other parameters depending on your specific problem.<|endoftext|>

# Uploaded  model

- **Developed by:** Ramikan-BR
- **License:** apache-2.0
- **Finetuned from model :** unsloth/qwen2-0.5b-bnb-4bit

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)