render examples

Author: t-kalinowski

vignettes/examples/index.Rmd

@@ -1,7 +1,7 @@
 ---
 title: Keras examples
 output: rmarkdown::html_vignette
-date: 'Last Modified: 2023-11-30; Last Rendered: 2024-09-04'
+date: 'Last Modified: 2023-11-30; Last Rendered: 2025-01-23'
 vignette: >
   %\VignetteIndexEntry{Keras examples}
   %\VignetteEngine{knitr::rmarkdown}

vignettes/examples/nlp/neural_machine_translation_with_transformer.Rmd

@@ -583,7 +583,7 @@ transformer |> fit(train_ds, epochs = epochs,
 ```
 
 ```
-## 1297/1297 - 48s - 37ms/step - accuracy: 0.7463 - loss: 1.8009 - val_accuracy: 0.7657 - val_loss: 1.5766
+## 1297/1297 - 64s - 50ms/step - accuracy: 0.7709 - loss: 1.5752 - val_accuracy: 0.7788 - val_loss: 1.3699
 ```
 
 

vignettes/examples/nlp/text_classification_from_scratch.Rmd

@@ -27,7 +27,6 @@ word splitting & indexing.
 library(tensorflow, exclude = c("shape", "set_random_seed"))
 library(tfdatasets, exclude = "shape")
 library(keras3)
-use_virtualenv("r-keras")
 ```
 
 ## Load the data: IMDB movie review sentiment classification
@@ -402,11 +401,11 @@ model |> fit(train_ds, validation_data = val_ds, epochs = epochs)
 
 ```
 ## Epoch 1/3
-## 625/625 - 6s - 10ms/step - accuracy: 0.6953 - loss: 0.5231 - val_accuracy: 0.8618 - val_loss: 0.3205
+## 625/625 - 5s - 8ms/step - accuracy: 0.6873 - loss: 0.5318 - val_accuracy: 0.8640 - val_loss: 0.3301
 ## Epoch 2/3
-## 625/625 - 1s - 2ms/step - accuracy: 0.9032 - loss: 0.2390 - val_accuracy: 0.8742 - val_loss: 0.3113
+## 625/625 - 2s - 3ms/step - accuracy: 0.9047 - loss: 0.2398 - val_accuracy: 0.8716 - val_loss: 0.3212
 ## Epoch 3/3
-## 625/625 - 2s - 2ms/step - accuracy: 0.9553 - loss: 0.1211 - val_accuracy: 0.8666 - val_loss: 0.3531
+## 625/625 - 2s - 3ms/step - accuracy: 0.9568 - loss: 0.1230 - val_accuracy: 0.8730 - val_loss: 0.3835
 ```
 
 ## Evaluate the model on the test set
@@ -417,15 +416,15 @@ model |> evaluate(test_ds)
 ```
 
 ```
-## 782/782 - 1s - 2ms/step - accuracy: 0.8520 - loss: 0.3960
+## 782/782 - 1s - 2ms/step - accuracy: 0.8655 - loss: 0.3982
 ```
 
 ```
 ## $accuracy
-## [1] 0.85204
+## [1] 0.86552
 ##
 ## $loss
-## [1] 0.3959631
+## [1] 0.3982031
 ```
 
 ## Make an end-to-end model
@@ -455,13 +454,13 @@ end_to_end_model |> evaluate(raw_test_ds)
 ```
 
 ```
-## 782/782 - 3s - 4ms/step - accuracy: 0.8520 - loss: 0.0000e+00
+## 782/782 - 3s - 4ms/step - accuracy: 0.8655 - loss: 0.3982
 ```
 
 ```
 ## $accuracy
-## [1] 0.85204
+## [1] 0.86552
 ##
 ## $loss
-## [1] 0
+## [1] 0.398203
 ```

vignettes/examples/structured_data/imbalanced_classification.Rmd

@@ -201,20 +201,20 @@ cat("Target: "); str(y)
 
 ```
 ## Input: List of 13
-##  $ age     :<tf.Tensor: shape=(), dtype=int32, numpy=45>
+##  $ age     :<tf.Tensor: shape=(), dtype=int32, numpy=59>
 ##  $ sex     :<tf.Tensor: shape=(), dtype=int32, numpy=1>
-##  $ cp      :<tf.Tensor: shape=(), dtype=int32, numpy=1>
-##  $ trestbps:<tf.Tensor: shape=(), dtype=int32, numpy=110>
-##  $ chol    :<tf.Tensor: shape=(), dtype=int32, numpy=264>
-##  $ fbs     :<tf.Tensor: shape=(), dtype=int32, numpy=0>
-##  $ restecg :<tf.Tensor: shape=(), dtype=int32, numpy=0>
-##  $ thalach :<tf.Tensor: shape=(), dtype=int32, numpy=132>
+##  $ cp      :<tf.Tensor: shape=(), dtype=int32, numpy=4>
+##  $ trestbps:<tf.Tensor: shape=(), dtype=int32, numpy=164>
+##  $ chol    :<tf.Tensor: shape=(), dtype=int32, numpy=176>
+##  $ fbs     :<tf.Tensor: shape=(), dtype=int32, numpy=1>
+##  $ restecg :<tf.Tensor: shape=(), dtype=int32, numpy=2>
+##  $ thalach :<tf.Tensor: shape=(), dtype=int32, numpy=90>
 ##  $ exang   :<tf.Tensor: shape=(), dtype=int32, numpy=0>
-##  $ oldpeak :<tf.Tensor: shape=(), dtype=float32, numpy=1.2>
+##  $ oldpeak :<tf.Tensor: shape=(), dtype=float32, numpy=1.0>
 ##  $ slope   :<tf.Tensor: shape=(), dtype=int32, numpy=2>
-##  $ ca      :<tf.Tensor: shape=(), dtype=int32, numpy=0>
-##  $ thal    :<tf.Tensor: shape=(), dtype=string, numpy=b'reversible'>
-## Target: <tf.Tensor: shape=(), dtype=int32, numpy=0>
+##  $ ca      :<tf.Tensor: shape=(), dtype=int32, numpy=2>
+##  $ thal    :<tf.Tensor: shape=(), dtype=string, numpy=b'fixed'>
+## Target: <tf.Tensor: shape=(), dtype=int32, numpy=1>
 ```
 
 Let's batch the datasets:
@@ -371,13 +371,13 @@ preprocessed_x
 
 ```
 ## tf.Tensor(
-## [[0. 0. 0. ... 0. 0. 0.]
+## [[0. 0. 0. ... 1. 0. 0.]
 ##  [0. 0. 0. ... 0. 0. 0.]
 ##  [0. 0. 0. ... 0. 0. 0.]
 ##  ...
 ##  [0. 0. 0. ... 0. 0. 0.]
 ##  [0. 0. 0. ... 0. 0. 0.]
-##  [0. 0. 0. ... 0. 0. 0.]], shape=(32, 136), dtype=float32)
+##  [0. 0. 1. ... 0. 0. 0.]], shape=(32, 136), dtype=float32)
 ```
 
 ## Two ways to manage preprocessing: as part of the `tf.data` pipeline, or in the model itself
@@ -463,45 +463,45 @@ training_model |> fit(
 
 ```
 ## Epoch 1/20
-## 8/8 - 2s - 280ms/step - accuracy: 0.4689 - loss: 0.7471 - val_accuracy: 0.5167 - val_loss: 0.7019
+## 8/8 - 2s - 274ms/step - accuracy: 0.4647 - loss: 0.7443 - val_accuracy: 0.4833 - val_loss: 0.7008
 ## Epoch 2/20
-## 8/8 - 1s - 140ms/step - accuracy: 0.5602 - loss: 0.6785 - val_accuracy: 0.6333 - val_loss: 0.6491
+## 8/8 - 0s - 29ms/step - accuracy: 0.6141 - loss: 0.6784 - val_accuracy: 0.6167 - val_loss: 0.6540
 ## Epoch 3/20
-## 8/8 - 0s - 46ms/step - accuracy: 0.6307 - loss: 0.6478 - val_accuracy: 0.7000 - val_loss: 0.6053
+## 8/8 - 0s - 32ms/step - accuracy: 0.5809 - loss: 0.6657 - val_accuracy: 0.7167 - val_loss: 0.6160
 ## Epoch 4/20
-## 8/8 - 0s - 12ms/step - accuracy: 0.6432 - loss: 0.6246 - val_accuracy: 0.7667 - val_loss: 0.5692
+## 8/8 - 0s - 30ms/step - accuracy: 0.6763 - loss: 0.6155 - val_accuracy: 0.7333 - val_loss: 0.5833
 ## Epoch 5/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7178 - loss: 0.5813 - val_accuracy: 0.7667 - val_loss: 0.5359
+## 8/8 - 0s - 30ms/step - accuracy: 0.7386 - loss: 0.5935 - val_accuracy: 0.7500 - val_loss: 0.5565
 ## Epoch 6/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7344 - loss: 0.5371 - val_accuracy: 0.7833 - val_loss: 0.5067
+## 8/8 - 0s - 30ms/step - accuracy: 0.7261 - loss: 0.5560 - val_accuracy: 0.7500 - val_loss: 0.5304
 ## Epoch 7/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7884 - loss: 0.5158 - val_accuracy: 0.8333 - val_loss: 0.4810
+## 8/8 - 0s - 30ms/step - accuracy: 0.7718 - loss: 0.5114 - val_accuracy: 0.7333 - val_loss: 0.5076
 ## Epoch 8/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7759 - loss: 0.5011 - val_accuracy: 0.8500 - val_loss: 0.4569
+## 8/8 - 0s - 30ms/step - accuracy: 0.7925 - loss: 0.5025 - val_accuracy: 0.7500 - val_loss: 0.4875
 ## Epoch 9/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7676 - loss: 0.4865 - val_accuracy: 0.8500 - val_loss: 0.4354
+## 8/8 - 0s - 30ms/step - accuracy: 0.7510 - loss: 0.5042 - val_accuracy: 0.7500 - val_loss: 0.4698
 ## Epoch 10/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7925 - loss: 0.4601 - val_accuracy: 0.8333 - val_loss: 0.4161
+## 8/8 - 0s - 30ms/step - accuracy: 0.8008 - loss: 0.4562 - val_accuracy: 0.7500 - val_loss: 0.4555
 ## Epoch 11/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7967 - loss: 0.4617 - val_accuracy: 0.8667 - val_loss: 0.3976
+## 8/8 - 0s - 30ms/step - accuracy: 0.8714 - loss: 0.4418 - val_accuracy: 0.7667 - val_loss: 0.4431
 ## Epoch 12/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.7967 - loss: 0.4316 - val_accuracy: 0.8667 - val_loss: 0.3796
+## 8/8 - 0s - 33ms/step - accuracy: 0.8506 - loss: 0.4182 - val_accuracy: 0.7500 - val_loss: 0.4327
 ## Epoch 13/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.8506 - loss: 0.4058 - val_accuracy: 0.8833 - val_loss: 0.3643
+## 8/8 - 0s - 30ms/step - accuracy: 0.8465 - loss: 0.3950 - val_accuracy: 0.7667 - val_loss: 0.4239
 ## Epoch 14/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.8174 - loss: 0.4197 - val_accuracy: 0.8833 - val_loss: 0.3510
+## 8/8 - 0s - 30ms/step - accuracy: 0.8382 - loss: 0.3905 - val_accuracy: 0.7667 - val_loss: 0.4166
 ## Epoch 15/20
-## 8/8 - 0s - 14ms/step - accuracy: 0.8299 - loss: 0.3888 - val_accuracy: 0.8833 - val_loss: 0.3405
+## 8/8 - 0s - 30ms/step - accuracy: 0.8465 - loss: 0.3661 - val_accuracy: 0.7833 - val_loss: 0.4104
 ## Epoch 16/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.8257 - loss: 0.3820 - val_accuracy: 0.8833 - val_loss: 0.3294
+## 8/8 - 0s - 30ms/step - accuracy: 0.8631 - loss: 0.3725 - val_accuracy: 0.8000 - val_loss: 0.4053
 ## Epoch 17/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.8299 - loss: 0.3746 - val_accuracy: 0.8833 - val_loss: 0.3223
+## 8/8 - 0s - 30ms/step - accuracy: 0.8299 - loss: 0.3679 - val_accuracy: 0.8167 - val_loss: 0.4014
 ## Epoch 18/20
-## 8/8 - 0s - 13ms/step - accuracy: 0.8506 - loss: 0.3487 - val_accuracy: 0.8833 - val_loss: 0.3153
+## 8/8 - 0s - 30ms/step - accuracy: 0.8714 - loss: 0.3501 - val_accuracy: 0.8167 - val_loss: 0.3984
 ## Epoch 19/20
-## 8/8 - 0s - 14ms/step - accuracy: 0.8465 - loss: 0.3558 - val_accuracy: 0.8667 - val_loss: 0.3093
+## 8/8 - 0s - 30ms/step - accuracy: 0.8714 - loss: 0.3322 - val_accuracy: 0.8167 - val_loss: 0.3949
 ## Epoch 20/20
-## 8/8 - 0s - 14ms/step - accuracy: 0.8672 - loss: 0.3570 - val_accuracy: 0.8667 - val_loss: 0.3036
+## 8/8 - 0s - 30ms/step - accuracy: 0.8506 - loss: 0.3303 - val_accuracy: 0.8167 - val_loss: 0.3924
 ```
 
 We quickly get to 80% validation accuracy.
@@ -534,7 +534,7 @@ predictions <- inference_model |> predict(input_dict)
 ```
 
 ```
-## 1/1 - 0s - 394ms/step
+## 1/1 - 0s - 341ms/step
 ```
 
 ``` r
@@ -545,6 +545,6 @@ glue::glue(r"---(
 ```
 
 ```
-## This particular patient had a 44.8% probability
+## This particular patient had a 51.4% probability
 ## of having a heart disease, as evaluated by our model.
 ```