Objective of the repository is to learn and build ML/DL models from scratch using Pytorch.
- Tensor basic: Tensor_Pytorch.ipynb
- Autograd: Autograd.ipynb
- Backpropagation: Backpropagation.ipynb
- Gradient_Descent_using_autograd: Gradient_Descent_using_autograd.ipynb
- Template pytorch: template.ipynb
- Feedforward neural network: feedforward_neural_network.ipynb
- Logistic regression: Logistic_regression.ipynb
- Linear regression: Linear_Regression.ipynb
- Writting custom dataset dataloader transform imgaug: custom_dataset_dataloader_transform_imgaug.ipynb
- Linear Regression: LinearRegression.ipynb
- k-nearest neighbors: KNN.ipynb
- Logistic Regression: LogisticRegression.ipynb
| CNN | Code | Reference |
|---|---|---|
| LeNet5 | LeNet_5.ipynb | images |
| AlexNet | AlexNet.ipynb | paper |
| VGG | VGG.ipynb | paper |
| InceptionResnet_v2 | InceptionResnet_v2.ipynb | paper |
| Multioutput CNN | Age_Gender_pipeline.ipynb | |
| ArcFace + fNet | arcface.ipynb |
| Model | Code | Reference |
|---|---|---|
| Vision Transformers | ViT.ipynb | [paper] [video] |
| Vision Transformers patch_4x2 | ViT_patch_4x2.ipynb | ... |
| Context-Cluster | Context-Cluster.ipynb | paper |
- LeNet: pytorch_lightning/LeNet.ipynb
- Multioutput CNN: Age_Gender_pipeline_pytorch_lightning.ipynb
- resnext50_32x4d_ArcFace: Pipeline_Image_Retrieval.ipynb
- Byte_Pair_Encoding: Byte_Pair_Encoding.ipynb
- Byte_Pair_Encoding Dropout: BPE_Dropout.ipynb
- Word2vec: Word_embedding_word2vec.ipynb
- LSTM: LSTM.ipynb