

Project 'VGG16_1FC.Exp02.BonB.100p.unmasked.fold_3':
[1] "Model type 'VGG16_1FC' sucessfully initialized:"
$model
Model
Model: "sequential"
____________________________________________________________________________________________________________________________________________________________________
Layer (type)                                                             Output Shape                                                      Param #                  
====================================================================================================================================================================
vgg16 (Model)                                                            (None, 7, 7, 512)                                                 14714688                 
____________________________________________________________________________________________________________________________________________________________________
flatten (Flatten)                                                        (None, 25088)                                                     0                        
____________________________________________________________________________________________________________________________________________________________________
fc1 (Dense)                                                              (None, 256)                                                       6422784                  
____________________________________________________________________________________________________________________________________________________________________
Predictions (Dense)                                                      (None, 10)                                                        2570                     
====================================================================================================================================================================
Total params: 21,140,042
Trainable params: 6,425,354
Non-trainable params: 14,714,688
____________________________________________________________________________________________________________________________________________________________________



$classes
                                     name id
1                Asteromphalus.labId_6835  0
2                  Chaetoceros.labId_6813  1
3  Fragilariopsis kerguelensis.labId_8356  2
4      Fragilariopsis rhombica.labId_8362  3
5                    Nitzschia.labId_6758  4
6              Pseudonitzschia.labId_8364  5
7                 Rhizosolenia.labId_6776  6
8            Silicoflagellate.labId_10255  7
9      Thalassiosira gracilis.labId_10366  8
10  Thalassiosira lentiginosa.labId_10369  9

$inputShape
[1] 224 224

$optimizer
<tensorflow.python.keras.optimizers.Adam>

$callbacksList
list()

Data:
Classes: 10

Training: 1164 samples:
   ClassId                              ClassName Count Count.ANT31 Count.PS103
7        0               Asteromphalus.labId_6835    96           0          96
4        1                 Chaetoceros.labId_6813   211           0         211
3        2 Fragilariopsis kerguelensis.labId_8356   145           0         145
8        3     Fragilariopsis rhombica.labId_8362   129           0         129
10       4                   Nitzschia.labId_6758    25           0          25
6        5             Pseudonitzschia.labId_8364   208           0         208
5        6                Rhizosolenia.labId_6776    44           0          44
9        7           Silicoflagellate.labId_10255    99           0          99
2        8     Thalassiosira gracilis.labId_10366    74           0          74
1        9  Thalassiosira lentiginosa.labId_10369   133           0         133
11      NA                                    Sum  1164           0        1164

Validation: 297 samples:
   ClassId                              ClassName Count Count.ANT31 Count.PS103
6        0               Asteromphalus.labId_6835    24           0          24
3        1                 Chaetoceros.labId_6813    53           0          53
5        2 Fragilariopsis kerguelensis.labId_8356    37           0          37
4        3     Fragilariopsis rhombica.labId_8362    33           0          33
10       4                   Nitzschia.labId_6758     7           0           7
2        5             Pseudonitzschia.labId_8364    53           0          53
8        6                Rhizosolenia.labId_6776    12           0          12
9        7           Silicoflagellate.labId_10255    25           0          25
7        8     Thalassiosira gracilis.labId_10366    19           0          19
1        9  Thalassiosira lentiginosa.labId_10369    34           0          34
11      NA                                    Sum   297           0         297

Test: 482 samples:
   ClassId                              ClassName Count Count.ANT31 Count.PS103
9        0               Asteromphalus.labId_6835    40           0          40
1        1                 Chaetoceros.labId_6813    88           0          88
2        2 Fragilariopsis kerguelensis.labId_8356    60           0          60
6        3     Fragilariopsis rhombica.labId_8362    53           0          53
10       4                   Nitzschia.labId_6758    10           0          10
4        5             Pseudonitzschia.labId_8364    86           0          86
7        6                Rhizosolenia.labId_6776    18           0          18
8        7           Silicoflagellate.labId_10255    41           0          41
3        8     Thalassiosira gracilis.labId_10366    31           0          31
5        9  Thalassiosira lentiginosa.labId_10369    55           0          55
11      NA                                    Sum   482           0         482

Compiling Model:

Starting Training:
Epochs: 50
Batch size: 32


Evaluating trained model for project 'VGG16_1FC.Exp02.BonB.100p.unmasked.fold_3':


Model evaluation:
$confusionMatrix
Confusion Matrix and Statistics

          Reference
Prediction  0  1  2  3  4  5  6  7  8  9
         0 39  0  0  0  0  0  0  0  0  0
         1  0 84  0  0  0  1  0  0  0  0
         2  0  0 60  0  0  0  0  0  0  0
         3  0  2  0 53  0  6  0  0  0  0
         4  0  0  0  0 10  1  0  0  0  0
         5  1  0  0  0  0 74  0  0  0  0
         6  0  1  0  0  0  3 18  0  0  0
         7  0  1  0  0  0  1  0 41  0  0
         8  0  0  0  0  0  0  0  0 31  0
         9  0  0  0  0  0  0  0  0  0 55

Overall Statistics
                                          
               Accuracy : 0.9647          
                 95% CI : (0.9441, 0.9793)
    No Information Rate : 0.1826          
    P-Value [Acc > NIR] : < 2.2e-16       
                                          
                  Kappa : 0.9598          
                                          
 Mcnemar's Test P-Value : NA              

Statistics by Class:

                     Class: 0 Class: 1 Class: 2 Class: 3 Class: 4 Class: 5 Class: 6 Class: 7 Class: 8 Class: 9
Precision             1.00000   0.9882   1.0000   0.8689  0.90909   0.9867  0.81818  0.95349  1.00000   1.0000
Recall                0.97500   0.9545   1.0000   1.0000  1.00000   0.8605  1.00000  1.00000  1.00000   1.0000
F1                    0.98734   0.9711   1.0000   0.9298  0.95238   0.9193  0.90000  0.97619  1.00000   1.0000
Prevalence            0.08299   0.1826   0.1245   0.1100  0.02075   0.1784  0.03734  0.08506  0.06432   0.1141
Detection Rate        0.08091   0.1743   0.1245   0.1100  0.02075   0.1535  0.03734  0.08506  0.06432   0.1141
Detection Prevalence  0.08091   0.1763   0.1245   0.1266  0.02282   0.1556  0.04564  0.08921  0.06432   0.1141
Balanced Accuracy     0.98750   0.9760   1.0000   0.9907  0.99894   0.9290  0.99569  0.99773  1.00000   1.0000

$statistics
   class                              className TP FP FN precision    recall        F1
1      0               Asteromphalus.labId_6835 39  0  1 1.0000000 0.9750000 0.9873418
2      1                 Chaetoceros.labId_6813 84  1  4 0.9882353 0.9545455 0.9710983
3      2 Fragilariopsis kerguelensis.labId_8356 60  0  0 1.0000000 1.0000000 1.0000000
4      3     Fragilariopsis rhombica.labId_8362 53  8  0 0.8688525 1.0000000 0.9298246
5      4                   Nitzschia.labId_6758 10  1  0 0.9090909 1.0000000 0.9523810
6      5             Pseudonitzschia.labId_8364 74  1 12 0.9866667 0.8604651 0.9192547
7      6                Rhizosolenia.labId_6776 18  4  0 0.8181818 1.0000000 0.9000000
8      7           Silicoflagellate.labId_10255 41  2  0 0.9534884 1.0000000 0.9761905
9      8     Thalassiosira gracilis.labId_10366 31  0  0 1.0000000 1.0000000 1.0000000
10     9  Thalassiosira lentiginosa.labId_10369 55  0  0 1.0000000 1.0000000 1.0000000

$macro
$macro$precision
[1] 0.9524516

$macro$recall
[1] 0.9790011

$macro$F1
[1] 0.9636091

$macro$F1.Sokolova_Lapalme
[1] 0.9655438


$micro
$micro$precision
[1] 0.9647303

$micro$recall
[1] 0.9647303

$micro$F1
[1] 0.9647303


