---
language: en
license: apache-2.0
tags:
  - image-classification
  - taxi
  - decoration
  - deep-learning
  - tensorflow
  - keras
---

# Taxi Decoration Prediction Model

This model predicts whether a taxi image is decorated or undecorated based on input image data. It was trained using a convolutional neural network (CNN) architecture on a custom dataset of decorated and undecorated taxi images.

# Taxi-Q-Learning Model

This is a Q-learning model trained to solve the Taxi-v3 environment. The model uses a **reinforcement learning** approach to optimize the agent's policy for navigating the taxi environment.

## Model Details

- **Model Type**: Q-Learning
- **Environment**: OpenAI Gym's `Taxi-v3`
- **Training Algorithm**: Q-learning
- **Input**: The state of the taxi environment, including the position of the taxi, destination, and status of the passenger.
- **Output**: Optimal action based on the current state.

## Intended Use
This model is intended to solve the Taxi-v3 environment, a simple reinforcement learning task where the goal is to pick up and drop off passengers at the correct locations.

### How to Use:
You can use this model for reinforcement learning tasks or to further train it in different environments.

```python
from huggingface_hub import hf_hub_download
import pickle

# Download the model
model_path = hf_hub_download(repo_id="willco-afk/taxi", filename="q-learning.pkl")

# Load the Q-learning model
with open(model_path, "rb") as f:
    q_learning_model = pickle.load(f)

# Use the model for your task