Kaggle SIIM-ISIC Melanoma Classification: EfficientNet-B7 Approach

---

Hardware / Software

Model training and predictions were performed in Kaggle TPU Kernels.

• Python 3.7.6 (Python packages are detailed separately in requirements.txt)
• Kaggle's TPU v3-8 (8 cores)

---

Data

• SIIM-ISIC Melanoma Classification
• ISIC 2019 TFRecords 256x256
• ISIC 2019 TFRecords 384x384
• ISIC 2019 TFRecords 512x512
• ISIC 2019 TFRecords 768x768
• Melanoma TFRecords 256x256
• Melanoma TFRecords 384x384
• Melanoma TFRecords 512x512
• Melanoma TFRecords 768x768
• melanoma_hairs

---

Image models

Training

We trained 8 image models, as shown below:

Model	Data	Image size	Epochs	Hair augmentation
1	2020	256x256	13
2	2020	384x384	15
3	2020	512x512	15
4	2020	768x768	15
5	2019-2020	256x256	25	X
6	2019-2020	384x384	25	X
7	2019-2020	512x512	12	X
8	2019-2020	768x768	10

To reproduce our results, the Kaggle notebook must be forked and executed 8 times, one for each model, changing only the content of the first cell (input) each time.

For example, for model 1, the content of the input cell should be:

tfrec_shape = 256
comp_data = "2020"

For model 6, however, it should be:

tfrec_shape = 384
comp_data = "2019-2020"

Training time for a single model ranges from ~ 0.5 to ~ 3 hours (all models can be fitted within 3 hours Kaggle's TPU limit for a single session).

Predicting

The source code for making predictions is published in this Kaggle notebook. It can also be found in "./kaggle_notebooks".

Again, to reproduce our results, the Kaggle notebook must be forked and executed 8 times, one for each model, changing only the content of the first cell (input) each time.

Our image-models submission files can be found in "./submissions/image_data".

---

Metadata

Training & Predicting

For metadata, I used a simple baseline model proposed by @titericz, which can be found here. The training model was adapted from rajnishe (thanks to @rajnishe) and ajaykumar7778 (thanks to @ajaykumar7778).