This is tutorial stuff I collected while looking into machinelearning
- https://github.com/glouw/tinn
- https://github.com/attractivechaos/kann
- https://github.com/codeplea/genann
- tensorflow
cd ./TensorFlow
python3 -m venv ./.venv
source ./.venv/bin/activate
pip3 install -r ./requirements.txt- Submodules
git submodule init
git submodule update- load dataset(s) (train data)
- build model
- compile model (optimizer, loss, metrics)
- train model (train data, epochs)
- [loss, acc] = evaluate model
- make prediction (raw data)
A model needs a loss function and an optimizer for training.
classification: select a class from a list regression: predict value
Loss is a measure of performance of a model. The lower, the better. When learning, the model aims to get the lowest loss possible. The target for multi-class classification is a one-hot vector, meaning it has 1 on a single position and 0’s everywhere else.
1. Define a question
2. Collect data
3. Visualize data
--> 4. Train algorithm
| 5. Test the Algorithm
| 6. Collect feedback
| 7. Refine the algorithm
--- 8. Loop 4-7 until the results are satisfying
9. Use the model to make a prediction
- https://github.com/glouw/tinn
- https://github.com/attractivechaos/kann
- https://github.com/codeplea/genann
- tensorflow
cd ./TensorFlow
python3 -m venv ./.venv
source ./.venv/bin/activate
pip3 install -r ./requirements.txt- Submodules
git submodule init
git submodule update- load dataset(s) (train data)
- build model
- compile model (optimizer, loss, metrics)
- train model (train data, epochs)
- [loss, acc] = evaluate model
- make prediction (raw data)
A model needs a loss function and an optimizer for training.
classification: select a class from a list regression: predict value
Loss is a measure of performance of a model. The lower, the better. When learning, the model aims to get the lowest loss possible. The target for multi-class classification is a one-hot vector, meaning it has 1 on a single position and 0’s everywhere else.
1. Define a question
2. Collect data
3. Visualize data
--> 4. Train algorithm
| 5. Test the Algorithm
| 6. Collect feedback
| 7. Refine the algorithm
--- 8. Loop 4-7 until the results are satisfying
9. Use the model to make a prediction