Deep Learning · TEE · guide
Numerical practice from direct to mixed
- Start here
- How to solve a mixed numerical without guessing
- Drill 1 Signs and brackets
- Drill 2 Half sum versus mean
- Drill 3 Natural log and probability
- Drill 4 Dot product and bias
- Drill 5 Matrix rows receive inputs
- Drill 6 Chain rule on a tiny loss
- Drill 7 Sigmoid with binary cross entropy
- Drill 8 Activation contrast
- Drill 9 Softmax and true class loss
- Drill 10 Sigmoid neuron gradient
- Drill 11 One perceptron learning update
- Drill 12 XOR using two hidden units
- Drill 13 Full ReLU backpropagation
- Drill 14 Two output sigmoid backpropagation
- Drill 15 A negative ReLU blocks this path
- Drill 16 Batch and sequential SGD
- Drill 17 Incomplete mini batch
- Drill 18 GD variation 1
- Drill 19 GD variation 2
- Drill 20 Momentum variation 1
- Drill 21 Momentum variation 2
- Drill 22 Nesterov variation 1
- Drill 23 Nesterov variation 2
- Drill 24 AdaGrad variation 1
- Drill 25 AdaGrad variation 2
- Drill 26 RMSProp variation 1
- Drill 27 RMSProp variation 2
- Drill 28 Adam variation 1
- Drill 29 Adam variation 2
- Drill 30 Regularized objective
- Drill 31 L1 update on positive and negative weights
- Drill 32 L2 update
- Drill 33 Inverted dropout
- Drill 34 Convolution with a negative coefficient
- Drill 35 Padding stride and channels
- Drill 36 Max and average pooling
- Drill 37 Complete small CNN
- Drill 38 Reverse a spatial formula
- Drill 39 Bottleneck cost with new values
- Drill 40 Residual addition
- Drill 41 Two RNN steps
- Drill 42 LSTM with a negative candidate
- Drill 43 Calculate gates before the cell
- Drill 44 GRU reset then mix
- Drill 45 BPTT shared weights
- Drill 46 Mixed convolution pool dense and loss
- Drill 47 Mixed backpropagation and L2
- Drill 48 Mixed dropout and a momentum update
- Drill 49 Mixed CNN dimensions softmax and loss
- Drill 50 Three step LSTM retention and replacement
- Drill 51 Inception concatenation and parameter count
- Drill 52 Projection shortcut
- Revising solved examples