fork download
  1. import numpy as np
  2. import tensorflow as tf
  3. # os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'
  4. # import keras
  5.  
  6. # Generate some example data (x values and corresponding y values)
  7. x_train = np.array([1, 2, 3, 4, 5], dtype=float)
  8. y_train = np.array([3, 5, 7, 9, 11], dtype=float) # y = 2x + 1
  9.  
  10. # Define a simple model
  11. model = tf.keras.Sequential([
  12. tf.keras.layers.Dense(units=1, input_shape=[1])
  13. ])
  14.  
  15. # Compile the model
  16. model.compile(optimizer='sgd', loss='mean_squared_error')
  17.  
  18. # Train the model
  19. model.fit(x_train, y_train, epochs=100)
  20.  
  21. # Predict
  22. print(model.predict([10])) # Should output something close to 21 (since y = 2*10 + 1)
Success #stdin #stdout #stderr 1.22s 217364KB
stdin
Standard input is empty
stdout
Epoch 1/100

5/5 [==============================] - 0s 27ms/sample - loss: 3.9200
Epoch 2/100

5/5 [==============================] - 0s 169us/sample - loss: 2.3382
Epoch 3/100

5/5 [==============================] - 0s 142us/sample - loss: 1.4161
Epoch 4/100

5/5 [==============================] - 0s 125us/sample - loss: 0.8784
Epoch 5/100

5/5 [==============================] - 0s 121us/sample - loss: 0.5646
Epoch 6/100

5/5 [==============================] - 0s 120us/sample - loss: 0.3814
Epoch 7/100

5/5 [==============================] - 0s 116us/sample - loss: 0.2743
Epoch 8/100

5/5 [==============================] - 0s 123us/sample - loss: 0.2116
Epoch 9/100

5/5 [==============================] - 0s 117us/sample - loss: 0.1747
Epoch 10/100

5/5 [==============================] - 0s 116us/sample - loss: 0.1528
Epoch 11/100

5/5 [==============================] - 0s 117us/sample - loss: 0.1397
Epoch 12/100

5/5 [==============================] - 0s 116us/sample - loss: 0.1317
Epoch 13/100

5/5 [==============================] - 0s 114us/sample - loss: 0.1268
Epoch 14/100

5/5 [==============================] - 0s 162us/sample - loss: 0.1235
Epoch 15/100

5/5 [==============================] - 0s 143us/sample - loss: 0.1213
Epoch 16/100

5/5 [==============================] - 0s 120us/sample - loss: 0.1197
Epoch 17/100

5/5 [==============================] - 0s 167us/sample - loss: 0.1184
Epoch 18/100

5/5 [==============================] - 0s 179us/sample - loss: 0.1173
Epoch 19/100

5/5 [==============================] - 0s 175us/sample - loss: 0.1164
Epoch 20/100

5/5 [==============================] - 0s 177us/sample - loss: 0.1155
Epoch 21/100

5/5 [==============================] - 0s 176us/sample - loss: 0.1146
Epoch 22/100

5/5 [==============================] - 0s 176us/sample - loss: 0.1138
Epoch 23/100

5/5 [==============================] - 0s 174us/sample - loss: 0.1130
Epoch 24/100

5/5 [==============================] - 0s 177us/sample - loss: 0.1123
Epoch 25/100

5/5 [==============================] - 0s 174us/sample - loss: 0.1115
Epoch 26/100

5/5 [==============================] - 0s 183us/sample - loss: 0.1108
Epoch 27/100

5/5 [==============================] - 0s 183us/sample - loss: 0.1100
Epoch 28/100

5/5 [==============================] - 0s 168us/sample - loss: 0.1093
Epoch 29/100

5/5 [==============================] - 0s 168us/sample - loss: 0.1085
Epoch 30/100

5/5 [==============================] - 0s 179us/sample - loss: 0.1078
Epoch 31/100

5/5 [==============================] - 0s 171us/sample - loss: 0.1071
Epoch 32/100

5/5 [==============================] - 0s 172us/sample - loss: 0.1063
Epoch 33/100

5/5 [==============================] - 0s 179us/sample - loss: 0.1056
Epoch 34/100

5/5 [==============================] - 0s 180us/sample - loss: 0.1049
Epoch 35/100

5/5 [==============================] - 0s 184us/sample - loss: 0.1042
Epoch 36/100

5/5 [==============================] - 0s 176us/sample - loss: 0.1035
Epoch 37/100

5/5 [==============================] - 0s 175us/sample - loss: 0.1028
Epoch 38/100

5/5 [==============================] - 0s 181us/sample - loss: 0.1021
Epoch 39/100

5/5 [==============================] - 0s 176us/sample - loss: 0.1014
Epoch 40/100

5/5 [==============================] - 0s 175us/sample - loss: 0.1007
Epoch 41/100

5/5 [==============================] - 0s 175us/sample - loss: 0.1000
Epoch 42/100

5/5 [==============================] - 0s 180us/sample - loss: 0.0994
Epoch 43/100

5/5 [==============================] - 0s 173us/sample - loss: 0.0987
Epoch 44/100

5/5 [==============================] - 0s 176us/sample - loss: 0.0980
Epoch 45/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0974
Epoch 46/100

5/5 [==============================] - 0s 176us/sample - loss: 0.0967
Epoch 47/100

5/5 [==============================] - 0s 174us/sample - loss: 0.0961
Epoch 48/100

5/5 [==============================] - 0s 177us/sample - loss: 0.0954
Epoch 49/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0948
Epoch 50/100

5/5 [==============================] - 0s 182us/sample - loss: 0.0941
Epoch 51/100

5/5 [==============================] - 0s 180us/sample - loss: 0.0935
Epoch 52/100

5/5 [==============================] - 0s 188us/sample - loss: 0.0929
Epoch 53/100

5/5 [==============================] - 0s 187us/sample - loss: 0.0922
Epoch 54/100

5/5 [==============================] - 0s 176us/sample - loss: 0.0916
Epoch 55/100

5/5 [==============================] - 0s 174us/sample - loss: 0.0910
Epoch 56/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0904
Epoch 57/100

5/5 [==============================] - 0s 175us/sample - loss: 0.0898
Epoch 58/100

5/5 [==============================] - 0s 173us/sample - loss: 0.0892
Epoch 59/100

5/5 [==============================] - 0s 180us/sample - loss: 0.0886
Epoch 60/100

5/5 [==============================] - 0s 177us/sample - loss: 0.0880
Epoch 61/100

5/5 [==============================] - 0s 185us/sample - loss: 0.0874
Epoch 62/100

5/5 [==============================] - 0s 181us/sample - loss: 0.0868
Epoch 63/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0862
Epoch 64/100

5/5 [==============================] - 0s 176us/sample - loss: 0.0856
Epoch 65/100

5/5 [==============================] - 0s 188us/sample - loss: 0.0850
Epoch 66/100

5/5 [==============================] - 0s 182us/sample - loss: 0.0845
Epoch 67/100

5/5 [==============================] - 0s 618us/sample - loss: 0.0839
Epoch 68/100

5/5 [==============================] - 0s 191us/sample - loss: 0.0833
Epoch 69/100

5/5 [==============================] - 0s 187us/sample - loss: 0.0828
Epoch 70/100

5/5 [==============================] - 0s 191us/sample - loss: 0.0822
Epoch 71/100

5/5 [==============================] - 0s 197us/sample - loss: 0.0816
Epoch 72/100

5/5 [==============================] - 0s 193us/sample - loss: 0.0811
Epoch 73/100

5/5 [==============================] - 0s 187us/sample - loss: 0.0806
Epoch 74/100

5/5 [==============================] - 0s 177us/sample - loss: 0.0800
Epoch 75/100

5/5 [==============================] - 0s 178us/sample - loss: 0.0795
Epoch 76/100

5/5 [==============================] - 0s 177us/sample - loss: 0.0789
Epoch 77/100

5/5 [==============================] - 0s 206us/sample - loss: 0.0784
Epoch 78/100

5/5 [==============================] - 0s 183us/sample - loss: 0.0779
Epoch 79/100

5/5 [==============================] - 0s 184us/sample - loss: 0.0773
Epoch 80/100

5/5 [==============================] - 0s 184us/sample - loss: 0.0768
Epoch 81/100

5/5 [==============================] - 0s 181us/sample - loss: 0.0763
Epoch 82/100

5/5 [==============================] - 0s 181us/sample - loss: 0.0758
Epoch 83/100

5/5 [==============================] - 0s 182us/sample - loss: 0.0753
Epoch 84/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0748
Epoch 85/100

5/5 [==============================] - 0s 176us/sample - loss: 0.0743
Epoch 86/100

5/5 [==============================] - 0s 177us/sample - loss: 0.0738
Epoch 87/100

5/5 [==============================] - 0s 172us/sample - loss: 0.0733
Epoch 88/100

5/5 [==============================] - 0s 177us/sample - loss: 0.0728
Epoch 89/100

5/5 [==============================] - 0s 172us/sample - loss: 0.0723
Epoch 90/100

5/5 [==============================] - 0s 171us/sample - loss: 0.0718
Epoch 91/100

5/5 [==============================] - 0s 170us/sample - loss: 0.0713
Epoch 92/100

5/5 [==============================] - 0s 184us/sample - loss: 0.0708
Epoch 93/100

5/5 [==============================] - 0s 176us/sample - loss: 0.0703
Epoch 94/100

5/5 [==============================] - 0s 172us/sample - loss: 0.0699
Epoch 95/100

5/5 [==============================] - 0s 172us/sample - loss: 0.0694
Epoch 96/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0689
Epoch 97/100

5/5 [==============================] - 0s 175us/sample - loss: 0.0685
Epoch 98/100

5/5 [==============================] - 0s 179us/sample - loss: 0.0680
Epoch 99/100

5/5 [==============================] - 0s 181us/sample - loss: 0.0675
Epoch 100/100

5/5 [==============================] - 0s 184us/sample - loss: 0.0671
[[22.07085]]
stderr
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/keras/utils/losses_utils.py:170: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.
WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.