---
license: mit
datasets:
- blanchon/EuroSAT_RGB
metrics:
- accuracy type:accuracy value:.88
library_name: transformers
language:
- en
pipeline_tag: image-classification
---


## Training Details

### Training Data
This model was trained on the Eurosat dataset containing  Sentinel-2 satellite images available at ```blanchon/EuroSAT_RGB```
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
The Eurosat dataset consists of ten classes and the a total of 27,000 images with a training set size of 16,200 images
  - Annual Crop
  - Forest
  - Herbaceous Vegetation
  - Highway
  - Industrial Buildings
  - Pasture
  - Permanent Crop
  - Residential Buildings
  - River
  - SeaLake


### Training Procedure

- Batch size: 24
- Optimizer: AdanW
- Learning Rate: 1e-4
- Criterion: CrossEntropyLoss
- Number of Epochs: 120
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->




#### Training Hyperparameters

- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->




## Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

### Testing Data, Factors & Metrics

#### Testing Data

- 5400 images

#### Metrics
Model Accuracy: 88%
model Recall: 88%
<!-- These are the evaluation metrics being used, ideally with a description of why. -->

[More Information Needed]

### Results

<img src="test.png" alt="CMatrix" width="400"/>


#### Summary