diff --git a/README.md b/README.md index 1ec49d2..4a70ebb 100644 --- a/README.md +++ b/README.md @@ -21,6 +21,10 @@ In addition, some models can be evaluated directly through [Akida Cloud](https:/ |---------------|-------------------|---------------------------------------------------------------------------|------------|---------------|---------|--------------|----------|----------|-------|----------------| | Vision | Classification | [MobileNetV1_1.0](vision/classification/mobilenetv1/cifar10) | 128 | CIFAR-10 | 2.25M | 8 | 91.92% | | | 5 ☁️ | | Vision | Classification | [MobileNetV1_1.0](vision/classification/mobilenetv1/oxford_flowers) | 224 | Oxford_Flower | 3.3M | 8 | 91.08% | | | 7 | +| Vision | Classification | [MobileNetV1 0.5](vision/classification/mobilenetv1/melanoma) | 224 | SIIM-ISIC | 3.14M | 8 | 98.16% | 86.14% | | 4 ☁️ | +| Vision | Classification | [MobileNetV1 0.5](vision/classification/mobilenetv1/odir-5k) | 224 | ODIR-5K | 3.15M | 8 | 92.00% | 97.83% | | 4 ☁️ | +| Vision | Classification | [MobileNetV1 0.5](vision/classification/mobilenetv1/ecg) | 224 | ECG | 3.14M | 8 | 83.27% | 89.18% | | 4 ☁️ | +| Vision | Classification | [MobileNetV1 0.5](vision/classification/mobilenetv1/retina_oct) | 224 | Retina OCT | 3.15M | 8 | 93.30% | 98.66% | | 4 ☁️ | | Vision | Classification | [MobileNetV2 1.0](vision/classification/mobilenetv2/imagenet) | 224 | ImageNet | 3.5M | 8 | 70.35% | | | 7 | | Vision | Classification | [MobileNetV2 0.75](vision/classification/mobilenetv2/imagenet) | 160 | ImageNet | 2.6M | 8 | 62.85% | | | 4 ☁️ | | Vision | Classification | [MobileNetV2 0.35](vision/classification/mobilenetv2/imagenet) | 96 | ImageNet | 1.2M | 8 | 43.47% | | | 2 ☁️ | diff --git a/vision/classification/mobilenetv1/ecg/README.md b/vision/classification/mobilenetv1/ecg/README.md new file mode 100644 index 0000000..3d0abe6 --- /dev/null +++ b/vision/classification/mobilenetv1/ecg/README.md @@ -0,0 +1,20 @@ +# MobileNetV1 (TF-Keras Model) + +## Source +This model uses the pre-trained base model [MobileNet from](tf_keras.applications.MobileNet) +- **Backbone**: MobileNetV1 pretrained on ImageNet +- **Alpha**: `MBV1_ALPHA = 0.5` + +## Environment +``` +tf_keras: 2.19 +cnn2snn: 2.17.0 +quantizeml: 1.0.1 +``` + +## References +- **MobileNets**: [Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) +- **Dataset**: [Physionet2017 ECG classification](https://www.kaggle.com/datasets/luigisaetta/physionet2017ecg) + +## License +Apache 2.0 License (models only) \ No newline at end of file diff --git a/vision/classification/mobilenetv1/ecg/mobilenet_ecg.h5 b/vision/classification/mobilenetv1/ecg/mobilenet_ecg.h5 new file mode 100644 index 0000000..afeffa1 --- /dev/null +++ b/vision/classification/mobilenetv1/ecg/mobilenet_ecg.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3d4a500eadf837dbcdd30384c30e1d78b16fb8f51e5c10210b8d43da8ff97ad5 +size 3589200 diff --git a/vision/classification/mobilenetv1/ecg/mobilenet_ecg_i8_w8_a8.h5 b/vision/classification/mobilenetv1/ecg/mobilenet_ecg_i8_w8_a8.h5 new file mode 100644 index 0000000..ad153a0 --- /dev/null +++ b/vision/classification/mobilenetv1/ecg/mobilenet_ecg_i8_w8_a8.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:99d34bf98c039966c6a6b981aba0d8b0e62f4c495e3a5df58151ff2c732515ce +size 3532872 diff --git a/vision/classification/mobilenetv1/melanoma/README.md b/vision/classification/mobilenetv1/melanoma/README.md new file mode 100644 index 0000000..5b5142a --- /dev/null +++ b/vision/classification/mobilenetv1/melanoma/README.md @@ -0,0 +1,20 @@ +# MobileNetV1 (TF-Keras Model) + +## Source +This model uses the pre-trained base model [MobileNet from](tf_keras.applications.MobileNet) +- **Backbone**: MobileNetV1 pretrained on ImageNet +- **Alpha**: `MBV1_ALPHA = 0.5` + +## Environment +``` +tf_keras: 2.19 +cnn2snn: 2.17.0 +quantizeml: 1.0.1 +``` + +## References +- **MobileNets**: [Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) +- **Dataset**: [SIIM-ISIC Melanoma Classification](https://challenge2020.isic-archive.com/) + +## License +Apache 2.0 License (models only) \ No newline at end of file diff --git a/vision/classification/mobilenetv1/melanoma/mobilenet_melanoma.h5 b/vision/classification/mobilenetv1/melanoma/mobilenet_melanoma.h5 new file mode 100644 index 0000000..c5bb8e9 --- /dev/null +++ b/vision/classification/mobilenetv1/melanoma/mobilenet_melanoma.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:51cbc21feffcbf9a7da991fb4699786e6c9ef699cbe60f90c8f9b90a64404f1e +size 3587136 diff --git a/vision/classification/mobilenetv1/melanoma/mobilenet_melanoma_i8_w8_a8.h5 b/vision/classification/mobilenetv1/melanoma/mobilenet_melanoma_i8_w8_a8.h5 new file mode 100644 index 0000000..b8dce06 --- /dev/null +++ b/vision/classification/mobilenetv1/melanoma/mobilenet_melanoma_i8_w8_a8.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:836e7f96df6a6b76c613e599eb0830ef55044b75469bd4ed26cd5beb71c98c08 +size 3530816 diff --git a/vision/classification/mobilenetv1/odir-5k/README.md b/vision/classification/mobilenetv1/odir-5k/README.md new file mode 100644 index 0000000..e65cfdf --- /dev/null +++ b/vision/classification/mobilenetv1/odir-5k/README.md @@ -0,0 +1,20 @@ +# MobileNetV1 (TF-Keras Model) + +## Source +This model uses the pre-trained base model [MobileNet from](tf_keras.applications.MobileNet) +- **Backbone**: MobileNetV1 pretrained on ImageNet +- **Alpha**: `MBV1_ALPHA = 0.5` + +## Environment +``` +tf_keras: 2.19 +cnn2snn: 2.17.0 +quantizeml: 1.0.1 +``` + +## References +- **MobileNets**: [Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) +- **Dataset**: [ODIR-5K Ocular disease recognition](https://www.kaggle.com/datasets/andrewmvd/ocular-disease-recognition-odir5k) + +## License +Apache 2.0 License (models only) \ No newline at end of file diff --git a/vision/classification/mobilenetv1/odir-5k/mobilenet_odir-5k.h5 b/vision/classification/mobilenetv1/odir-5k/mobilenet_odir-5k.h5 new file mode 100644 index 0000000..099fd40 --- /dev/null +++ b/vision/classification/mobilenetv1/odir-5k/mobilenet_odir-5k.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:067275d84c0f0d7fc4364cd79164863f7027eacf8d39f6f34a720c150ea46fd9 +size 3593280 diff --git a/vision/classification/mobilenetv1/odir-5k/mobilenet_odir-5k_i8_w8_a8.h5 b/vision/classification/mobilenetv1/odir-5k/mobilenet_odir-5k_i8_w8_a8.h5 new file mode 100644 index 0000000..b30831d --- /dev/null +++ b/vision/classification/mobilenetv1/odir-5k/mobilenet_odir-5k_i8_w8_a8.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d948a714722232a77889be5af57f953776c658a9939715b9378cb6d069c17692 +size 3536960 diff --git a/vision/classification/mobilenetv1/retina_oct/README.md b/vision/classification/mobilenetv1/retina_oct/README.md new file mode 100644 index 0000000..139aee4 --- /dev/null +++ b/vision/classification/mobilenetv1/retina_oct/README.md @@ -0,0 +1,20 @@ +# MobileNetV1 (TF-Keras Model) + +## Source +This model uses the pre-trained base model [MobileNet from](tf_keras.applications.MobileNet) +- **Backbone**: MobileNetV1 pretrained on ImageNet +- **Alpha**: `MBV1_ALPHA = 0.5` + +## Environment +``` +tf_keras: 2.19 +cnn2snn: 2.17.0 +quantizeml: 1.0.1 +``` + +## References +- **MobileNets**: [Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) +- **Dataset**: [Retinal OCT ocular disease recognition](https://www.kaggle.com/datasets/paultimothymooney/kermany2018) + +## License +Apache 2.0 License (models only) \ No newline at end of file diff --git a/vision/classification/mobilenetv1/retina_oct/mobilenet_retina_oct.h5 b/vision/classification/mobilenetv1/retina_oct/mobilenet_retina_oct.h5 new file mode 100644 index 0000000..447df4b --- /dev/null +++ b/vision/classification/mobilenetv1/retina_oct/mobilenet_retina_oct.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ba49dcc97af99c3ceba282036e639f13cc06626e7d46cd251f5760ee2c9bcfd +size 3593304 diff --git a/vision/classification/mobilenetv1/retina_oct/mobilenet_retina_oct_i8_w8_a8.h5 b/vision/classification/mobilenetv1/retina_oct/mobilenet_retina_oct_i8_w8_a8.h5 new file mode 100644 index 0000000..108776c --- /dev/null +++ b/vision/classification/mobilenetv1/retina_oct/mobilenet_retina_oct_i8_w8_a8.h5 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3931c17534c8391cfae19d017e264f3390de4da6a5baaef70e2e3c0aa8d50088 +size 3536992