---
library_name: litert
pipeline_tag: image-classification
tags:
  - vision
  - image-classification
  - computer-vision
datasets:
  - imagenet-1k
model-index:
  - name: inception_v3
    results:
      - task:
          type: image-classification
          name: Image Classification
        dataset:
          name: ImageNet-1k
          type: imagenet-1k
          config: default
          split: validation
        metrics:
          - name: Top 1 Accuracy (Full Precision)
            type: accuracy
            value: 0.7727
          - name: Top 5 Accuracy (Full Precision)
            type: accuracy
            value: 0.9343
---

# Inception_v3

Inception v3 model pre-trained on ImageNet-1k. It was introduced in [**Rethinking the Inception Architecture for Computer Vision**](https://arxiv.org/abs/1512.00567) by Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, and Zbigniew Wojna.

## Intended uses & limitations

The model files were converted from pretrained weights from PyTorch Vision. The models may have their own licenses or terms and conditions derived from PyTorch Vision and the dataset used for training. It is your responsibility to determine whether you have permission to use the models for your use case.

## Model description

The model was converted from a checkpoint from PyTorch Vision [`Inception_V3_Weights.IMAGENET1K_V1`](https://docs.pytorch.org/vision/main/models/generated/torchvision.models.inception_v3.html#torchvision.models.Inception_V3_Weights). 

The original model has:    
acc@1 (on ImageNet-1K): 77.294%    
acc@5 (on ImageNet-1K): 93.450%    
num_params: 27,161,264   

## Use
```python
#!/usr/bin/env python3
import argparse, json
import numpy as np
from PIL import Image
from huggingface_hub import hf_hub_download
from ai_edge_litert.compiled_model import CompiledModel

def preprocess(img: Image.Image) -> np.ndarray:
    img = img.convert("RGB")
    w, h = img.size
    # Inception_v3 expects a resize to 342 prior to the 299 central crop
    s = 342
    if w < h:
        img = img.resize((s, int(round(h * s / w))), Image.BILINEAR)
    else:
        img = img.resize((int(round(w * s / h)), s), Image.BILINEAR)
        
    # Central crop to 299x299
    left = (img.size[0] - 299) // 2
    top = (img.size[1] - 299) // 2
    img = img.crop((left, top, left + 299, top + 299))
    
    # Rescale to [0.0, 1.0] and Normalize
    x = np.asarray(img, dtype=np.float32) / 255.0
    x = (x - np.array([0.485, 0.456, 0.406], dtype=np.float32)) / np.array(
        [0.229, 0.224, 0.225], dtype=np.float32
    )
    # Expand dimensions to create NHWC 4D tensor: (1, 299, 299, 3)
    x = np.expand_dims(x, axis=0)

    return x

def main():
    ap = argparse.ArgumentParser()
    ap.add_argument("--image", required=True)
    args = ap.parse_args()

    # Download the TFLite model and labels
    model_path = hf_hub_download("litert-community/inception_v3", "inception_v3.tflite")
    labels_path = hf_hub_download(
        "huggingface/label-files", "imagenet-1k-id2label.json", repo_type="dataset"
    )
    
    with open(labels_path, "r", encoding="utf-8") as f:
        id2label = {int(k): v for k, v in json.load(f).items()}

    img = Image.open(args.image)
    x = preprocess(img)

    model = CompiledModel.from_file(model_path)
    inp = model.create_input_buffers(0)
    out = model.create_output_buffers(0)

    inp[0].write(x)
    model.run_by_index(0, inp, out)

    req = model.get_output_buffer_requirements(0, 0)
    y = out[0].read(req["buffer_size"] // np.dtype(np.float32).itemsize, np.float32)

    pred = int(np.argmax(y))
    label = id2label.get(pred, f"class_{pred}")

    print(f"Top-1 class index: {pred}")
    print(f"Top-1 label: {label}")

if __name__ == "__main__":
    main()

```
### BibTeX entry and citation info

```bibtex
@inproceedings{szegedy2016rethinking,
  title={Rethinking the inception architecture for computer vision},
  author={Szegedy, Christian and Vanhoucke, Vincent and Ioffe, Sergey and Shlens, Jon and Wojna, Zbigniew},
  booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
  pages={2818--2826},
  year={2016}
}
```