- Dataset is provided by Kaggle uploaded by Meinertsen
- Link to File : https://www.kaggle.com/meinertsen/new-york-city-taxi-trip-hourly-weather-data
- The dataset can be fetch using kaggle library on colab
- But i also provided the extracted dataset (weather.csv)
- 10.400 samples
- 80:20 train test split
- Tensorflow
- Keras
- Scikit learn
- matplotlib
- Numpy
- Pandas
- Variable used: tempm (hourly mean temperature)
- Data Shape : (10476, 2)
- Check Null
- Drop NUll (5 rows)
- Check Dtypes\
- (Variable maximum value - variable minimum value) * (10/100)
- 0.2 validation set
- Shuffle False (time series data is sensitive to shuffled data)
- set Random state
- 2 hidden layer (perceptron 512 units and 128 units)
- output layer 'Softmax'
- model = keras.Sequential([
- layers.Conv1D(filters=32, kernel_size=5,
-
strides=1, padding="causal",
-
activation="relu",
-
input_shape=[None, 1]),
- layers.LSTM(64, return_sequences=True),
- layers.LSTM(64, return_sequences=True),
- layers.Dense(30, activation="relu"),
- layers.Dense(10, activation="relu"),
- layers.Dense(1),
- layers.Lambda(lambda x: x * 400)
- ])
- loss : Huber
- optimizer Sgd with learning rate
- metrics MAE
- Set to stop training if val accuracy is smaller than target MAE
- epoch : 500
- callbacks
-
Best Validation Loss: 4.91
-
Best Validation MAE: 5.37
-
Loss
- Accuracy