---
license: apache-2.0
datasets:
- christykoh/imdb_pt
language:
- pt
metrics:
- accuracy
library_name: transformers
pipeline_tag: text-classification
tags:
- sentiment-analysis
widget:
- text: "Esqueceram de mim 2 é um dos melhores filmes de natal de todos os tempos."
  example_title: Exemplo
- text: "Esqueceram de mim 2 é o pior filme da franquia inteira."
  example_title: Exemplo
---
# TeenyTinyLlama-162m-IMDB

TeenyTinyLlama is a series of small foundational models trained on Portuguese.

This repository contains a version of [TeenyTinyLlama-162m](https://huggingface.co/nicholasKluge/TeenyTinyLlama-162m) fine-tuned on a translated version of the [IMDB dataset](https://huggingface.co/datasets/christykoh/imdb_pt).

## Reproducing
  
```python
# IMDB
! pip install transformers datasets evaluate accelerate -q

import evaluate
import numpy as np
from datasets import load_dataset
from transformers import AutoTokenizer, DataCollatorWithPadding
from transformers import AutoModelForSequenceClassification, TrainingArguments, Trainer

# Load the task
dataset = load_dataset("christykoh/imdb_pt")

# Create a `ModelForSequenceClassification`
model = AutoModelForSequenceClassification.from_pretrained(
    "nicholasKluge/TeenyTinyLlama-162m", 
    num_labels=2, 
    id2label={0: "NEGATIVE", 1: "POSITIVE"}, 
    label2id={"NEGATIVE": 0, "POSITIVE": 1}
)

tokenizer = AutoTokenizer.from_pretrained("nicholasKluge/TeenyTinyLlama-162m")

# Preprocess the dataset
def preprocess_function(examples):
    return tokenizer(examples["text"], truncation=True, max_length=256)

dataset_tokenized = dataset.map(preprocess_function, batched=True)

# Create a simple data collactor
data_collator = DataCollatorWithPadding(tokenizer=tokenizer)

# Use accuracy as an evaluation metric
accuracy = evaluate.load("accuracy")

# Function to compute accuracy
def compute_metrics(eval_pred):
    predictions, labels = eval_pred
    predictions = np.argmax(predictions, axis=1)
    return accuracy.compute(predictions=predictions, references=labels)

# Define training arguments
training_args = TrainingArguments(
    output_dir="checkpoints",
    learning_rate=4e-5,
    per_device_train_batch_size=16,
    per_device_eval_batch_size=16,
    num_train_epochs=3,
    weight_decay=0.01,
    evaluation_strategy="epoch",
    save_strategy="epoch",
    load_best_model_at_end=True,
    push_to_hub=True,
    hub_token="your_token_here",
    hub_model_id="username/model-name-imdb"
)

# Define the Trainer
trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=dataset_tokenized["train"],
    eval_dataset=dataset_tokenized["test"],
    tokenizer=tokenizer,
    data_collator=data_collator,
    compute_metrics=compute_metrics,
)

# Train!
trainer.train()
```

## Results

| Models                                                                                     | [IMDB](https://huggingface.co/datasets/christykoh/imdb_pt) | 
|--------------------------------------------------------------------------------------------|------------------------------------------------------------|
| [Teeny Tiny Llama 162m](https://huggingface.co/nicholasKluge/TeenyTinyLlama-162m)          | 91.14                                                      |
| [Bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) | 92.22                                                      |
| [Gpt2-small-portuguese](https://huggingface.co/pierreguillou/gpt2-small-portuguese)        | 91.60                                                      |