render guides

Author: t-kalinowski

vignettes/custom_train_step_in_tensorflow.Rmd

@@ -127,7 +127,7 @@ model |> fit(x, y, epochs = 3)
 
 ```
 ## Epoch 1/3
-## 32/32 - 1s - 24ms/step - mae: 1.4339 - loss: 3.2271
+## 32/32 - 1s - 29ms/step - mae: 1.4339 - loss: 3.2271
 ## Epoch 2/3
 ## 32/32 - 0s - 2ms/step - mae: 1.3605 - loss: 2.9034
 ## Epoch 3/3
@@ -212,11 +212,11 @@ model |> fit(x, y, epochs = 3)
 
 ```
 ## Epoch 1/3
-## 32/32 - 1s - 23ms/step - loss: 2.5170 - mae: 1.2923
+## 32/32 - 1s - 20ms/step - loss: 2.5170 - mae: 1.2923
 ## Epoch 2/3
 ## 32/32 - 0s - 2ms/step - loss: 2.2689 - mae: 1.2241
 ## Epoch 3/3
-## 32/32 - 0s - 3ms/step - loss: 2.0578 - mae: 1.1633
+## 32/32 - 0s - 2ms/step - loss: 2.0578 - mae: 1.1633
 ```
 
 ## Supporting `sample_weight` & `class_weight`
@@ -282,11 +282,11 @@ model |> fit(x, y, sample_weight = sw, epochs = 3)
 
 ```
 ## Epoch 1/3
-## 32/32 - 1s - 26ms/step - mae: 1.3434 - loss: 0.1681
+## 32/32 - 1s - 28ms/step - mae: 1.3434 - loss: 0.1681
 ## Epoch 2/3
-## 32/32 - 0s - 2ms/step - mae: 1.3364 - loss: 0.1394
+## 32/32 - 0s - 3ms/step - mae: 1.3364 - loss: 0.1394
 ## Epoch 3/3
-## 32/32 - 0s - 2ms/step - mae: 1.3286 - loss: 0.1148
+## 32/32 - 0s - 3ms/step - mae: 1.3286 - loss: 0.1148
 ```
 
 ## Providing your own evaluation step
@@ -332,7 +332,7 @@ model |> evaluate(x, y)
 ```
 
 ```
-## 32/32 - 0s - 10ms/step - mae: 1.3871 - loss: 0.0000e+00
+## 32/32 - 0s - 9ms/step - mae: 1.3871 - loss: 0.0000e+00
 ```
 
 ```
@@ -508,7 +508,7 @@ gan |> fit(
 ```
 
 ```
-## 100/100 - 5s - 55ms/step - d_loss: 0.0000e+00 - g_loss: 0.0000e+00
+## 100/100 - 6s - 57ms/step - d_loss: 0.0000e+00 - g_loss: 0.0000e+00
 ```
 
 The ideas behind deep learning are simple, so why should their implementation be painful?

vignettes/distribution.Rmd

@@ -184,24 +184,24 @@ model |> fit(dataset, epochs = 3)
 
 ```
 ## Epoch 1/3
-## 8/8 - 0s - 40ms/step - loss: 1.1398
+## 8/8 - 0s - 38ms/step - loss: 1.1533
 ## Epoch 2/3
-## 8/8 - 0s - 7ms/step - loss: 1.0593
+## 8/8 - 0s - 5ms/step - loss: 1.0621
 ## Epoch 3/3
-## 8/8 - 0s - 5ms/step - loss: 1.0071
+## 8/8 - 0s - 7ms/step - loss: 1.0163
 ```
 
 ``` r
 model |> evaluate(dataset)
 ```
 
 ```
-## 8/8 - 0s - 6ms/step - loss: 0.9609
+## 8/8 - 0s - 7ms/step - loss: 0.9673
 ```
 
 ```
 ## $loss
-## [1] 0.9609067
+## [1] 0.9673058
 ```
 
 
@@ -278,24 +278,24 @@ model |> fit(dataset, epochs = 3)
 
 ```
 ## Epoch 1/3
-## 8/8 - 0s - 25ms/step - loss: 1.1612
+## 8/8 - 0s - 42ms/step - loss: 1.1424
 ## Epoch 2/3
-## 8/8 - 0s - 4ms/step - loss: 1.0702
+## 8/8 - 0s - 7ms/step - loss: 1.0528
 ## Epoch 3/3
-## 8/8 - 0s - 5ms/step - loss: 1.0184
+## 8/8 - 0s - 7ms/step - loss: 1.0393
 ```
 
 ``` r
 model |> evaluate(dataset)
 ```
 
 ```
-## 8/8 - 0s - 7ms/step - loss: 0.9844
+## 8/8 - 0s - 9ms/step - loss: 1.0088
 ```
 
 ```
 ## $loss
-## [1] 0.9844129
+## [1] 1.008847
 ```
 
 

vignettes/functional_api.Rmd

@@ -208,17 +208,17 @@ history <- model |> fit(
 
 ```
 ## Epoch 1/2
-## 750/750 - 2s - 3ms/step - accuracy: 0.8980 - loss: 0.3540 - val_accuracy: 0.9444 - val_loss: 0.1898
+## 750/750 - 2s - 3ms/step - accuracy: 0.8979 - loss: 0.3540 - val_accuracy: 0.9448 - val_loss: 0.1903
 ## Epoch 2/2
-## 750/750 - 1s - 1ms/step - accuracy: 0.9512 - loss: 0.1633 - val_accuracy: 0.9605 - val_loss: 0.1387
+## 750/750 - 1s - 2ms/step - accuracy: 0.9509 - loss: 0.1635 - val_accuracy: 0.9597 - val_loss: 0.1397
 ```
 
 ``` r
 test_scores <- model |> evaluate(x_test, y_test, verbose=2)
 ```
 
 ```
-## 313/313 - 1s - 2ms/step - accuracy: 0.9598 - loss: 0.1322
+## 313/313 - 1s - 2ms/step - accuracy: 0.9595 - loss: 0.1328
 ```
 
 
@@ -229,8 +229,8 @@ cat("Test accuracy:", test_scores$accuracy, "\n")
 ```
 
 ```
-## Test loss: 0.1321879
-## Test accuracy: 0.9598
+## Test loss: 0.132778
+## Test accuracy: 0.9595
 ```
 
 For further reading, see the [training and evaluation](training_with_built_in_methods.html) guide.
@@ -643,7 +643,7 @@ model |> fit(
 
 ```
 ## Epoch 1/2
-## 40/40 - 3s - 65ms/step - department_loss: 2.8465 - loss: 0.7669 - priority_loss: 0.1976
+## 40/40 - 3s - 74ms/step - department_loss: 2.8465 - loss: 0.7669 - priority_loss: 0.1976
 ## Epoch 2/2
 ## 40/40 - 0s - 6ms/step - department_loss: 2.8554 - loss: 0.7538 - priority_loss: 0.1828
 ```
@@ -776,7 +776,7 @@ model |> fit(
 ```
 
 ```
-## 13/13 - 5s - 367ms/step - acc: 0.1250 - loss: 2.3003 - val_acc: 0.1250 - val_loss: 2.2946
+## 13/13 - 6s - 453ms/step - acc: 0.1250 - loss: 2.3006 - val_acc: 0.1300 - val_loss: 2.2969
 ```
 
 ## Shared layers

vignettes/intro_to_keras_for_engineers.Rmd

@@ -178,25 +178,25 @@ model |> fit(
 
 ```
 ## Epoch 1/10
-## 399/399 - 8s - 20ms/step - acc: 0.7476 - loss: 0.7467 - val_acc: 0.9663 - val_loss: 0.1179
+## 399/399 - 7s - 18ms/step - acc: 0.7496 - loss: 0.7385 - val_acc: 0.9641 - val_loss: 0.1228
 ## Epoch 2/10
-## 399/399 - 2s - 5ms/step - acc: 0.9384 - loss: 0.2066 - val_acc: 0.9770 - val_loss: 0.0765
+## 399/399 - 3s - 7ms/step - acc: 0.9382 - loss: 0.2037 - val_acc: 0.9769 - val_loss: 0.0774
 ## Epoch 3/10
-## 399/399 - 2s - 5ms/step - acc: 0.9569 - loss: 0.1467 - val_acc: 0.9817 - val_loss: 0.0622
+## 399/399 - 3s - 7ms/step - acc: 0.9567 - loss: 0.1458 - val_acc: 0.9816 - val_loss: 0.0636
 ## Epoch 4/10
-## 399/399 - 2s - 5ms/step - acc: 0.9652 - loss: 0.1170 - val_acc: 0.9860 - val_loss: 0.0499
+## 399/399 - 3s - 7ms/step - acc: 0.9658 - loss: 0.1163 - val_acc: 0.9866 - val_loss: 0.0468
 ## Epoch 5/10
-## 399/399 - 2s - 5ms/step - acc: 0.9709 - loss: 0.0999 - val_acc: 0.9873 - val_loss: 0.0447
+## 399/399 - 3s - 9ms/step - acc: 0.9719 - loss: 0.0975 - val_acc: 0.9880 - val_loss: 0.0433
 ## Epoch 6/10
-## 399/399 - 2s - 5ms/step - acc: 0.9752 - loss: 0.0863 - val_acc: 0.9877 - val_loss: 0.0400
+## 399/399 - 3s - 8ms/step - acc: 0.9758 - loss: 0.0853 - val_acc: 0.9874 - val_loss: 0.0413
 ## Epoch 7/10
-## 399/399 - 2s - 5ms/step - acc: 0.9764 - loss: 0.0787 - val_acc: 0.9890 - val_loss: 0.0395
+## 399/399 - 3s - 7ms/step - acc: 0.9765 - loss: 0.0782 - val_acc: 0.9891 - val_loss: 0.0398
 ## Epoch 8/10
-## 399/399 - 2s - 5ms/step - acc: 0.9794 - loss: 0.0678 - val_acc: 0.9874 - val_loss: 0.0432
+## 399/399 - 3s - 8ms/step - acc: 0.9797 - loss: 0.0678 - val_acc: 0.9881 - val_loss: 0.0419
 ## Epoch 9/10
-## 399/399 - 2s - 5ms/step - acc: 0.9802 - loss: 0.0658 - val_acc: 0.9894 - val_loss: 0.0395
+## 399/399 - 3s - 7ms/step - acc: 0.9805 - loss: 0.0652 - val_acc: 0.9897 - val_loss: 0.0381
 ## Epoch 10/10
-## 399/399 - 2s - 5ms/step - acc: 0.9825 - loss: 0.0584 - val_acc: 0.9914 - val_loss: 0.0342
+## 399/399 - 3s - 8ms/step - acc: 0.9831 - loss: 0.0576 - val_acc: 0.9912 - val_loss: 0.0340
 ```
 
 ``` r
@@ -227,7 +227,7 @@ predictions <- model |> predict(x_test)
 ```
 
 ```
-## 313/313 - 0s - 2ms/step
+## 313/313 - 1s - 2ms/step
 ```
 
 ``` r
@@ -362,7 +362,7 @@ model |> fit(
 ```
 
 ```
-## 399/399 - 6s - 15ms/step - acc: 0.7343 - loss: 0.7741 - val_acc: 0.9269 - val_loss: 0.2399
+## 399/399 - 7s - 18ms/step - acc: 0.7344 - loss: 0.7749 - val_acc: 0.9259 - val_loss: 0.2411
 ```
 
 ## Training models on arbitrary data sources
@@ -439,7 +439,7 @@ model |> fit(train_dataset, epochs = 1, validation_data = test_dataset)
 ```
 
 ```
-## 469/469 - 7s - 14ms/step - acc: 0.7499 - loss: 0.7454 - val_acc: 0.9051 - val_loss: 0.3089
+## 469/469 - 8s - 17ms/step - acc: 0.7493 - loss: 0.7476 - val_acc: 0.9123 - val_loss: 0.2965
 ```
 
 ## Further reading

vignettes/making_new_layers_and_models_via_subclassing.Rmd

@@ -96,10 +96,11 @@ linear_layer$weights
 
 ```
 ## [[1]]
-## <KerasVariable shape=(2, 4), dtype=float32, path=linear/variable>
+## <Variable path=linear/variable, shape=(2, 4), dtype=float32, value=[[-0.06251299  0.05335509  0.01485647 -0.00985784]
+##  [ 0.08404355  0.10115016  0.00569303  0.05479009]]>
 ##
 ## [[2]]
-## <KerasVariable shape=(4), dtype=float32, path=linear/variable_1>
+## <Variable path=linear/variable_1, shape=(4), dtype=float32, value=[0. 0. 0. 0.]>
 ```
 
 ## Layers can have non-trainable weights
@@ -480,7 +481,7 @@ model |> fit(random_normal(c(2, 3)), random_normal(c(2, 3)), epochs = 1)
 ```
 
 ```
-## 1/1 - 0s - 144ms/step - loss: 1.9081
+## 1/1 - 0s - 142ms/step - loss: 1.9081
 ```
 
 ``` r
@@ -492,7 +493,7 @@ model |> fit(random_normal(c(2, 3)), random_normal(c(2, 3)), epochs = 1)
 ```
 
 ```
-## 1/1 - 0s - 78ms/step - loss: -2.2532e-03
+## 1/1 - 0s - 115ms/step - loss: 1.6613
 ```
 
 ## You can optionally enable serialization on your layers
@@ -841,7 +842,7 @@ vae |> fit(x_train, x_train, epochs = 2, batch_size = 64)
 
 ```
 ## Epoch 1/2
-## 938/938 - 4s - 4ms/step - loss: 0.0748
+## 938/938 - 5s - 5ms/step - loss: 0.0748
 ## Epoch 2/2
-## 938/938 - 1s - 810us/step - loss: 0.0676
+## 938/938 - 1s - 2ms/step - loss: 0.0676
 ```

vignettes/sequential_model.Rmd

@@ -167,10 +167,13 @@ layer$weights  # Now it has weights, of shape (4, 3) and (3,)
 
 ```
 ## [[1]]
-## <KerasVariable shape=(4, 3), dtype=float32, path=dense_9/kernel>
+## <Variable path=dense_9/kernel, shape=(4, 3), dtype=float32, value=[[ 0.48581433  0.78749573  0.61015   ]
+##  [ 0.7962619   0.7261175  -0.8046875 ]
+##  [-0.6189915   0.37973273  0.50559556]
+##  [-0.5455791  -0.60714126  0.19791973]]>
 ##
 ## [[2]]
-## <KerasVariable shape=(3), dtype=float32, path=dense_9/bias>
+## <Variable path=dense_9/bias, shape=(3), dtype=float32, value=[0. 0. 0.]>
 ```
 
 Naturally, this also applies to Sequential models. When you instantiate a