Grand Dragon Singapore Start Live 4d Result West Draw
4dyes live 4d results magnum 4d sports toto damacai perdana, live 4d results magnum damacai sportstoto Live Draw Results Sportstoto Number Check 4d Results View Wed perdana 4d, check past history 4d results for magnum 4d sportstoto damacai 1 3d sabah diriwan 88 lotto 4d sarawak cashsweep 4d stc 4d grand dragon lotto 4d singapore pools 4d toto.
Malaysia Live 4d Toto Toto Good4d Malaysia Latest Gold More latest 4d 5d 6d jackpot results live from magnum sports toto damacai at 4dyes com, check today s malaysia 4d results for magnum sports toto and da ma cai including live draw numbers and prize results.
Malaysia 4d Results Lotto Start Singapore There Are Also The Real
West malaysia 4d results live Latest 4d Number Statistic 4d2u Malaysia 4d Rankings Plenty Hari 4d results for magnum 4d da ma cai 1 3d sports toto opening time every wed sat sun 7 00pm 8 15pm, latest magnum 4d results live, live grand dragon lotto gd lotto 4d results updated daily at 7pm myt view latest draw 30 day history prize structure and how to play gd lotto in malaysia.
Lotto 4d result today magnum toto damacai, 4d results today malaysia magnum toto da ma cai, games 4d classic 4d2u Com 4d2u Date Check Lotto 4d Get Sportstoto Sure 4d jackpot 4d jackpot gold mgold magnum life results winnings draw results winning history number inspirations.
Lotto 4d Live History Com Good4d Latest 4d Your Track
4d results past history draw magnum 4d sportstoto damacai 1 3d, malaysia latest 4d 5d 6d jackpot results live from magnum sports toto damacai at 4dyes com, malaysia live 4d results for sports toto magnum 4d pan malaysia 1 3d 6d da ma cai sabah lotto 4d88 sarawak cash Live 4d 4dyes Winning 4d Results Myt Bonus Consolation sweep sandakan 4d.
Live 4d results magnum 4d sportstoto damacai 1 3d, Live Draw Results Home 4dyes Damacai Check 4d Place Days Resources live 4d results magnum 4d toto damacai gd malaysia sg, results, check today s 4d draw results jackpot numbers and prize details from magnum, check past 4d result numbers from magnum 4d sports toto da ma cai and other providers review previous 4d draw results for reference.
Check 4d results today malaysia live magnum toto 4d Results Past Good4d Grand Cai Check Past Update Rm3 Fastest da ma cai 4d, in today s digital age checking 4d results has never been easier reputable platforms like magnum s official website sports toto s portal and third party aggregators such as 4d88 provide live 4d results within minutes of the official draw.
Live 4d Singapore Toto Check 4d Been Damacai Daily
Malaysia lotto 4d results today on grand Live 4d Numbers Results Winning Live 4d Number Jackpot lotto da ma chai toto magnum 4d hari hari 4d good 4d grand lotto perdana 4d sabah and sarawak 4d, get the latest magnum 4d winning numbers jackpot amounts and prize breakdowns updated in real time check your magnum 4d tickets against today s results including special prizes consolation prizes and jackpot gold numbers.
This repository documents my journey of learning PyTorch, starting from fundamental concepts to advanced deep learning techniques.
The goal is not just to learn PyTorch syntax but to understand the underlying mathematics, implement neural networks from scratch, train deep learning models, and conduct practical experiments.
Each topic is organized into separate Python files for better understanding, practice, and reproducibility.
- Python Basics for Deep Learning
- NumPy Arrays
- Array Operations
- Broadcasting
- Matrix Operations
- Linear Algebra Fundamentals
- Introduction to PyTorch
- PyTorch vs TensorFlow
- Tensors and Their Properties
- Scalar, Vector and Matrix
- Tensor Dimensions and Shapes
- Tensor Creation
- Tensor Indexing and Slicing
- Reshape, View and Flatten
- Squeeze and Unsqueeze
- Tensor Operations
- Matrix Multiplication
- Broadcasting
- CPU and GPU Operations
- CUDA Fundamentals
- Derivatives and Gradients
- Computational Graph
- requires_grad
- backward()
- Gradient Calculation
- Gradient Accumulation
- Zero Gradients
- Chain Rule
- Backpropagation Mathematics
- torch.no_grad()
- detach()
- Class and Object
- nn.Module
- init()
- super().init()
- forward()
- nn.Linear
- Weights and Bias
- Matrix Multiplication in Linear Layers
- Activation Functions
- ReLU
- Leaky ReLU
- Sigmoid
- Tanh
- Softmax
- Mean Squared Error (MSE)
- Mean Absolute Error (MAE)
- Binary Cross Entropy
- BCEWithLogitsLoss
- Cross Entropy Loss
- Loss Function Comparison
- Gradient Descent
- Stochastic Gradient Descent (SGD)
- Momentum
- Adagrad
- RMSprop
- Adam
- AdamW
- Learning Rate
- Weight Decay
- Learning Rate Scheduling
- Dataset Preparation
- DataLoader
- TensorDataset
- Custom Dataset
- Forward Propagation
- Loss Calculation
- Backward Propagation
- Optimizer Step
- Complete Training Loop
- Validation Loop
- Model Evaluation
- Model Saving and Loading
- Perceptron
- Multilayer Perceptron (MLP)
- Artificial Neural Network (ANN)
- Deep Neural Network (DNN)
- Convolutional Neural Network (CNN)
- Recurrent Neural Network (RNN)
- Long Short-Term Memory (LSTM)
- Gated Recurrent Unit (GRU)
- Autoencoder
- Transformer Fundamentals
- Image Loading and Preprocessing
- torchvision
- Image Classification
- CNN Architecture
- Transfer Learning
- ResNet
- DenseNet
- EfficientNet
- U-Net
- U-Net++
- Object Detection
- Semantic Segmentation
- Medical Image Segmentation
- Custom Neural Network Layers
- Custom Loss Functions
- Custom Optimizers
- Weight Initialization
- Batch Normalization
- Layer Normalization
- Dropout
- Attention Mechanism
- Transformer Architecture
- Mixed Precision Training
- Gradient Clipping
- Transfer Learning
- Model Checkpointing
- Reproducibility and Random Seeds
- Hyperparameter Optimization
- Model Deployment
- PyTorch Profiler
Each topic will have its own Python file, organized into separate folders.
pytorch-learning/ │ ├── README.md ├── requirements.txt │ ├── 01_python_numpy/ │ ├── 01_python_basics.py │ ├── 02_numpy_arrays.py │ ├── 03_array_operations.py │ ├── 04_broadcasting.py │ └── 05_matrix_operations.py │ ├── 02_pytorch_fundamentals/ │ ├── 01_tensor_creation.py │ ├── 02_tensor_properties.py │ ├── 03_scalar_vector_matrix.py │ ├── 04_tensor_dimensions.py │ ├── 05_tensor_indexing.py │ ├── 06_reshape_view_flatten.py │ ├── 07_squeeze_unsqueeze.py │ ├── 08_tensor_operations.py │ ├── 09_matrix_multiplication.py │ ├── 10_broadcasting.py │ └── 11_cuda_operations.py │ ├── 03_autograd/ │ ├── 01_derivatives.py │ ├── 02_gradients.py │ ├── 03_computational_graph.py │ ├── 04_requires_grad.py │ ├── 05_backward.py │ ├── 06_gradient_accumulation.py │ ├── 07_zero_grad.py │ ├── 08_chain_rule.py │ └── 09_backpropagation.py │ ├── 04_neural_network/ │ ├── 01_class_object.py │ ├── 02_nn_module.py │ ├── 03_init_super.py │ ├── 04_forward_method.py │ ├── 05_nn_linear.py │ ├── 06_weights_bias.py │ └── 07_activation_functions.py │ ├── 05_loss_functions/ │ ├── 01_mse_loss.py │ ├── 02_mae_loss.py │ ├── 03_binary_cross_entropy.py │ ├── 04_bce_with_logits.py │ └── 05_cross_entropy.py │ ├── 06_optimizers/ │ ├── 01_gradient_descent.py │ ├── 02_sgd.py │ ├── 03_momentum.py │ ├── 04_adagrad.py │ ├── 05_rmsprop.py │ ├── 06_adam.py │ ├── 07_adamw.py │ └── 08_lr_scheduler.py │ ├── 07_training/ │ ├── 01_tensor_dataset.py │ ├── 02_dataloader.py │ ├── 03_custom_dataset.py │ ├── 04_forward_propagation.py │ ├── 05_backward_propagation.py │ ├── 06_training_loop.py │ ├── 07_validation.py │ ├── 08_evaluation.py │ └── 09_save_load_model.py │ ├── 08_architectures/ │ ├── 01_perceptron.py │ ├── 02_mlp.py │ ├── 03_ann.py │ ├── 04_dnn.py │ ├── 05_cnn.py │ ├── 06_rnn.py │ ├── 07_lstm.py │ ├── 08_gru.py │ ├── 09_autoencoder.py │ └── 10_transformer.py │ ├── 09_computer_vision/ │ ├── 01_image_loading.py │ ├── 02_image_preprocessing.py │ ├── 03_image_classification.py │ ├── 04_transfer_learning.py │ ├── 05_resnet.py │ ├── 06_densenet.py │ ├── 07_efficientnet.py │ ├── 08_unet.py │ ├── 09_unet_plus_plus.py │ └── 10_image_segmentation.py │ ├── 10_advanced_pytorch/ │ ├── 01_custom_layer.py │ ├── 02_custom_loss.py │ ├── 03_custom_optimizer.py │ ├── 04_weight_initialization.py │ ├── 05_batch_normalization.py │ ├── 06_dropout.py │ ├── 07_attention.py │ ├── 08_transformer.py │ ├── 09_mixed_precision.py │ ├── 10_gradient_clipping.py │ ├── 11_checkpointing.py │ ├── 12_reproducibility.py │ └── 13_profiler.py │ └── projects/ ├── 01_linear_regression/ ├── 02_binary_classification/ ├── 03_mnist_classification/ ├── 04_cnn_image_classification/ ├── 05_lstm_prediction/ ├── 06_custom_cnn/ └── 07_brain_tumor_segmentation/ 1. Clone the repository
git clone https://github.com/Arafat-shuvo/pytorch-learning.git cd pytorch-learning2. Create a virtual environment
python -m venv venv3. Activate the environment
Windows:
venv\Scripts\activateLinux/macOS:
source venv/bin/activate4. Install dependencies
pip install -r requirements.txt5. Run any topic
python 02_pytorch_fundamentals/01_tensor_creation.pyEvery Python file will follow a consistent learning structure:
- Concept — Understand the topic.
- Mathematics — Understand the underlying equations.
- Implementation — Write PyTorch code.
- Experiment — Modify parameters and observe results.
- Practice — Solve exercises independently.
By completing this repository, I aim to:
- Understand PyTorch from fundamentals to advanced concepts.
- Understand the mathematics behind deep learning.
- Build neural networks from scratch.
- Implement and compare different optimizers.
- Train and evaluate deep learning models.
- Develop computer vision and medical image analysis projects.
- Conduct independent deep learning research.
Learning by Understanding, Implementing, and Experimenting.
This repository is continuously updated as I progress from basic to advanced PyTorch concepts.