---
license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
- Image Regression
datasets:
- "BrownEnergy/secchi_depth"
metrics:
- accuracy
model-index:
- name: "sd_depth_regression_v2"
  results: []
---

# sd_depth_regression_v2
## Image Regression Model

This model was trained with [Image Regression Model Trainer](https://github.com/TonyAssi/ImageRegression/tree/main). It takes an image as input and outputs a float value.

```python
from ImageRegression import predict
predict(repo_id='BrownEnergy/sd_depth_regression_v2',image_path='image.jpg')
```

---

## Dataset
Dataset: BrownEnergy/secchi_depth\
Value Column: 'sd_depth'\
Train Test Split: 0.05

---

## Training
Base Model: [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224)\
Epochs: 10\
Learning Rate: 0.0001

---

## Usage

### Download
```bash
git clone https://github.com/TonyAssi/ImageRegression.git
cd ImageRegression
```

### Installation
```bash
pip install -r requirements.txt
```

### Import 
```python
from ImageRegression import train_model, upload_model, predict
```

### Inference (Prediction)
- **repo_id** 🤗 repo id of the model
- **image_path** path to image
```python
predict(repo_id='BrownEnergy/sd_depth_regression_v2',
        image_path='image.jpg')
```
The first time this function is called it'll download the safetensor model. Subsequent function calls will run faster.

### Train Model
- **dataset_id** 🤗 dataset id
- **value_column_name** column name of prediction values in dataset
- **test_split** test split of the train/test split
- **output_dir** the directory where the checkpoints will be saved
- **num_train_epochs** training epochs
- **learning_rate** learning rate
```python
train_model(dataset_id='BrownEnergy/secchi_depth',
            value_column_name='sd_depth',
            test_split=0.05,
            output_dir='./results',
            num_train_epochs=10,
            learning_rate=0.0001)

```
The trainer will save the checkpoints in the output_dir location. The model.safetensors are the trained weights you'll use for inference (predicton).

### Upload Model
This function will upload your model to the 🤗 Hub.
- **model_id** the name of the model id
- **token** go [here](https://huggingface.co/settings/tokens) to create a new 🤗 token
- **checkpoint_dir** checkpoint folder that will be uploaded
```python
upload_model(model_id='sd_depth_regression_v2',
             token='YOUR_HF_TOKEN',
             checkpoint_dir='./results/checkpoint-940')
```