matrix-spam-ml

commit 6328a8b0e4fbeb8798c08ea94c5c1d259d655fa3
parent 500cd2923951e2a9ce670ad0eba0da06678b238b
Author: MTRNord <mtrnord1@gmail.com>
Date:   Sat, 10 Dec 2022 17:01:54 +0100

Add tests to bert model and print how many are correct in the end

Diffstat:
M
bert.ipynb
 | 
393
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++------------
M
model_v2.py
 | 
27
++++++++++++++++-----------

2 files changed, 351 insertions(+), 69 deletions(-)
diff --git a/bert.ipynb b/bert.ipynb
@@ -9,20 +9,15 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "2022-12-08 00:35:43.038652: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
+      "2022-12-10 16:51:06.512818: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
       "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
-      "2022-12-08 00:35:43.146406: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n"
+      "2022-12-10 16:51:06.619243: E tensorflow/stream_executor/cuda/cuda_blas.cc:2981] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "raw_train_ds: <TensorSliceDataset element_spec=(TensorSpec(shape=(), dtype=tf.string, name=None), TensorSpec(shape=(), dtype=tf.int64, name=None))>\n",
-      "train_ds: <PrefetchDataset element_spec=(TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None,), dtype=tf.int64, name=None))>\n",
-      "raw_val_ds: <TensorSliceDataset element_spec=(TensorSpec(shape=(), dtype=tf.string, name=None), TensorSpec(shape=(), dtype=tf.int64, name=None))>\n",
-      "Message: b'Not heard from U4 a while. Call me now am here all night with just my knickers on. Make me beg for it like U did last time 01223585236 XX Luv Nikiyu4.net'\n",
-      "Label: 1\n",
       "WARNING:tensorflow:Please fix your imports. Module tensorflow.python.training.tracking.data_structures has been moved to tensorflow.python.trackable.data_structures. The old module will be deleted in version 2.11.\n"
      ]
     },
@@ -30,18 +25,18 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "2022-12-08 00:35:44.682218: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:44.685968: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:44.686152: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:44.686543: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
+      "2022-12-10 16:51:08.213715: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.217539: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.217978: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.218526: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
       "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
-      "2022-12-08 00:35:44.687309: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:44.687473: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:44.687611: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:45.084063: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:45.084255: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:45.084410: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
-      "2022-12-08 00:35:45.084736: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1616] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 4245 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 2080 SUPER, pci bus id: 0000:08:00.0, compute capability: 7.5\n"
+      "2022-12-10 16:51:08.219226: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.219388: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.219528: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.626829: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.627024: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.627186: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:980] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
+      "2022-12-10 16:51:08.627323: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1616] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 4463 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 2080 SUPER, pci bus id: 0000:08:00.0, compute capability: 7.5\n"
      ]
     },
     {
@@ -54,18 +49,16 @@
       "==================================================================================================\n",
       " text (InputLayer)              [(None,)]            0           []                               \n",
       "                                                                                                  \n",
-      " preprocessing (KerasLayer)     {'input_type_ids':   0           ['text[0][0]']                   \n",
-      "                                (None, 128),                                                      \n",
+      " preprocessing (KerasLayer)     {'input_mask': (Non  0           ['text[0][0]']                   \n",
+      "                                e, 128),                                                          \n",
       "                                 'input_word_ids':                                                \n",
       "                                (None, 128),                                                      \n",
-      "                                 'input_mask': (Non                                               \n",
-      "                                e, 128)}                                                          \n",
+      "                                 'input_type_ids':                                                \n",
+      "                                (None, 128)}                                                      \n",
       "                                                                                                  \n",
-      " BERT_encoder (KerasLayer)      {'pooled_output': (  13548801    ['preprocessing[0][0]',          \n",
-      "                                None, 256),                       'preprocessing[0][1]',          \n",
-      "                                 'default': (None,                'preprocessing[0][2]']          \n",
-      "                                256),                                                             \n",
-      "                                 'encoder_outputs':                                               \n",
+      " BERT_encoder (KerasLayer)      {'sequence_output':  13548801    ['preprocessing[0][0]',          \n",
+      "                                 (None, 128, 256),                'preprocessing[0][1]',          \n",
+      "                                 'encoder_outputs':               'preprocessing[0][2]']          \n",
       "                                 [(None, 128, 256),                                               \n",
       "                                 (None, 128, 256),                                                \n",
       "                                 (None, 128, 256),                                                \n",
@@ -78,8 +71,10 @@
       "                                 (None, 128, 256),                                                \n",
       "                                 (None, 128, 256),                                                \n",
       "                                 (None, 128, 256)],                                               \n",
-      "                                 'sequence_output':                                               \n",
-      "                                 (None, 128, 256)}                                                \n",
+      "                                 'pooled_output': (                                               \n",
+      "                                None, 256),                                                       \n",
+      "                                 'default': (None,                                                \n",
+      "                                256)}                                                             \n",
       "                                                                                                  \n",
       " dropout (Dropout)              (None, 256)          0           ['BERT_encoder[0][13]']          \n",
       "                                                                                                  \n",
@@ -91,19 +86,109 @@
       "Non-trainable params: 1\n",
       "__________________________________________________________________________________________________\n",
       "Training model with https://tfhub.dev/google/electra_small/2\n",
-      "Epoch 1/5\n",
-      "42/42 [==============================] - 21s 254ms/step - loss: 0.3527 - binary_accuracy: 0.8231 - val_loss: 0.0972 - val_binary_accuracy: 0.9738\n",
-      "Epoch 2/5\n",
-      "42/42 [==============================] - 10s 240ms/step - loss: 0.0553 - binary_accuracy: 0.9790 - val_loss: 0.1025 - val_binary_accuracy: 0.9720\n",
-      "Epoch 3/5\n",
-      "42/42 [==============================] - 10s 239ms/step - loss: 0.0262 - binary_accuracy: 0.9918 - val_loss: 0.1065 - val_binary_accuracy: 0.9738\n",
-      "Epoch 4/5\n",
-      "42/42 [==============================] - 10s 236ms/step - loss: 0.0167 - binary_accuracy: 0.9940 - val_loss: 0.1009 - val_binary_accuracy: 0.9755\n",
-      "Epoch 5/5\n",
-      "42/42 [==============================] - 10s 236ms/step - loss: 0.0164 - binary_accuracy: 0.9948 - val_loss: 0.0980 - val_binary_accuracy: 0.9755\n",
-      "18/18 [==============================] - 1s 65ms/step - loss: 0.0980 - binary_accuracy: 0.9755\n",
-      "Loss: 0.09797481447458267\n",
-      "Accuracy: 0.9755244851112366\n",
+      "Epoch 1/50\n",
+      "42/42 [==============================] - 22s 254ms/step - loss: 0.6233 - binary_accuracy: 0.7166 - val_loss: 0.3777 - val_binary_accuracy: 0.8094\n",
+      "Epoch 2/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 0.2135 - binary_accuracy: 0.9190 - val_loss: 0.1298 - val_binary_accuracy: 0.9615\n",
+      "Epoch 3/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 0.0726 - binary_accuracy: 0.9715 - val_loss: 0.1039 - val_binary_accuracy: 0.9720\n",
+      "Epoch 4/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 0.0391 - binary_accuracy: 0.9858 - val_loss: 0.1421 - val_binary_accuracy: 0.9685\n",
+      "Epoch 5/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 0.0233 - binary_accuracy: 0.9933 - val_loss: 0.1572 - val_binary_accuracy: 0.9703\n",
+      "Epoch 6/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 0.0176 - binary_accuracy: 0.9910 - val_loss: 0.0614 - val_binary_accuracy: 0.9843\n",
+      "Epoch 7/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 0.0029 - binary_accuracy: 0.9993 - val_loss: 0.0675 - val_binary_accuracy: 0.9860\n",
+      "Epoch 8/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 0.0043 - binary_accuracy: 0.9985 - val_loss: 0.0694 - val_binary_accuracy: 0.9860\n",
+      "Epoch 9/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 8.5732e-04 - binary_accuracy: 1.0000 - val_loss: 0.0759 - val_binary_accuracy: 0.9843\n",
+      "Epoch 10/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 0.0014 - binary_accuracy: 0.9993 - val_loss: 0.0866 - val_binary_accuracy: 0.9860\n",
+      "Epoch 11/50\n",
+      "42/42 [==============================] - 10s 238ms/step - loss: 3.7205e-04 - binary_accuracy: 1.0000 - val_loss: 0.0820 - val_binary_accuracy: 0.9843\n",
+      "Epoch 12/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 2.6485e-04 - binary_accuracy: 1.0000 - val_loss: 0.0902 - val_binary_accuracy: 0.9843\n",
+      "Epoch 13/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 1.9019e-04 - binary_accuracy: 1.0000 - val_loss: 0.1003 - val_binary_accuracy: 0.9843\n",
+      "Epoch 14/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 1.7637e-04 - binary_accuracy: 1.0000 - val_loss: 0.1004 - val_binary_accuracy: 0.9843\n",
+      "Epoch 15/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 1.6374e-04 - binary_accuracy: 1.0000 - val_loss: 0.1018 - val_binary_accuracy: 0.9843\n",
+      "Epoch 16/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 1.3852e-04 - binary_accuracy: 1.0000 - val_loss: 0.1082 - val_binary_accuracy: 0.9808\n",
+      "Epoch 17/50\n",
+      "42/42 [==============================] - 10s 238ms/step - loss: 1.1150e-04 - binary_accuracy: 1.0000 - val_loss: 0.1099 - val_binary_accuracy: 0.9825\n",
+      "Epoch 18/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 1.0550e-04 - binary_accuracy: 1.0000 - val_loss: 0.1126 - val_binary_accuracy: 0.9808\n",
+      "Epoch 19/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 9.8560e-05 - binary_accuracy: 1.0000 - val_loss: 0.1117 - val_binary_accuracy: 0.9825\n",
+      "Epoch 20/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 2.8649e-04 - binary_accuracy: 1.0000 - val_loss: 0.1030 - val_binary_accuracy: 0.9808\n",
+      "Epoch 21/50\n",
+      "42/42 [==============================] - 10s 243ms/step - loss: 1.0976e-04 - binary_accuracy: 1.0000 - val_loss: 0.0909 - val_binary_accuracy: 0.9843\n",
+      "Epoch 22/50\n",
+      "42/42 [==============================] - 10s 242ms/step - loss: 1.1211e-04 - binary_accuracy: 1.0000 - val_loss: 0.1115 - val_binary_accuracy: 0.9825\n",
+      "Epoch 23/50\n",
+      "42/42 [==============================] - 10s 237ms/step - loss: 7.6401e-05 - binary_accuracy: 1.0000 - val_loss: 0.1225 - val_binary_accuracy: 0.9808\n",
+      "Epoch 24/50\n",
+      "42/42 [==============================] - 10s 230ms/step - loss: 7.7119e-05 - binary_accuracy: 1.0000 - val_loss: 0.1205 - val_binary_accuracy: 0.9808\n",
+      "Epoch 25/50\n",
+      "42/42 [==============================] - 10s 231ms/step - loss: 6.6323e-05 - binary_accuracy: 1.0000 - val_loss: 0.1215 - val_binary_accuracy: 0.9808\n",
+      "Epoch 26/50\n",
+      "42/42 [==============================] - 9s 227ms/step - loss: 6.6501e-05 - binary_accuracy: 1.0000 - val_loss: 0.1223 - val_binary_accuracy: 0.9808\n",
+      "Epoch 27/50\n",
+      "42/42 [==============================] - 9s 226ms/step - loss: 6.2443e-05 - binary_accuracy: 1.0000 - val_loss: 0.1284 - val_binary_accuracy: 0.9825\n",
+      "Epoch 28/50\n",
+      "42/42 [==============================] - 10s 232ms/step - loss: 6.8873e-05 - binary_accuracy: 1.0000 - val_loss: 0.1209 - val_binary_accuracy: 0.9808\n",
+      "Epoch 29/50\n",
+      "42/42 [==============================] - 10s 235ms/step - loss: 5.7817e-05 - binary_accuracy: 1.0000 - val_loss: 0.1201 - val_binary_accuracy: 0.9808\n",
+      "Epoch 30/50\n",
+      "42/42 [==============================] - 10s 234ms/step - loss: 6.2797e-05 - binary_accuracy: 1.0000 - val_loss: 0.1188 - val_binary_accuracy: 0.9808\n",
+      "Epoch 31/50\n",
+      "42/42 [==============================] - 10s 232ms/step - loss: 5.5777e-05 - binary_accuracy: 1.0000 - val_loss: 0.1165 - val_binary_accuracy: 0.9825\n",
+      "Epoch 32/50\n",
+      "42/42 [==============================] - 10s 234ms/step - loss: 5.4295e-05 - binary_accuracy: 1.0000 - val_loss: 0.1235 - val_binary_accuracy: 0.9808\n",
+      "Epoch 33/50\n",
+      "42/42 [==============================] - 10s 234ms/step - loss: 5.0356e-05 - binary_accuracy: 1.0000 - val_loss: 0.1228 - val_binary_accuracy: 0.9808\n",
+      "Epoch 34/50\n",
+      "42/42 [==============================] - 10s 235ms/step - loss: 4.6275e-05 - binary_accuracy: 1.0000 - val_loss: 0.1244 - val_binary_accuracy: 0.9808\n",
+      "Epoch 35/50\n",
+      "42/42 [==============================] - 10s 237ms/step - loss: 4.5953e-05 - binary_accuracy: 1.0000 - val_loss: 0.1258 - val_binary_accuracy: 0.9808\n",
+      "Epoch 36/50\n",
+      "42/42 [==============================] - 10s 233ms/step - loss: 4.3018e-05 - binary_accuracy: 1.0000 - val_loss: 0.1272 - val_binary_accuracy: 0.9808\n",
+      "Epoch 37/50\n",
+      "42/42 [==============================] - 10s 237ms/step - loss: 4.3228e-05 - binary_accuracy: 1.0000 - val_loss: 0.1282 - val_binary_accuracy: 0.9808\n",
+      "Epoch 38/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 4.2014e-05 - binary_accuracy: 1.0000 - val_loss: 0.1284 - val_binary_accuracy: 0.9808\n",
+      "Epoch 39/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 4.1043e-05 - binary_accuracy: 1.0000 - val_loss: 0.1283 - val_binary_accuracy: 0.9808\n",
+      "Epoch 40/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 4.7759e-05 - binary_accuracy: 1.0000 - val_loss: 0.1274 - val_binary_accuracy: 0.9808\n",
+      "Epoch 41/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 3.9910e-05 - binary_accuracy: 1.0000 - val_loss: 0.1296 - val_binary_accuracy: 0.9808\n",
+      "Epoch 42/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 4.3200e-05 - binary_accuracy: 1.0000 - val_loss: 0.1332 - val_binary_accuracy: 0.9808\n",
+      "Epoch 43/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 5.3371e-05 - binary_accuracy: 1.0000 - val_loss: 0.1274 - val_binary_accuracy: 0.9808\n",
+      "Epoch 44/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 3.8588e-05 - binary_accuracy: 1.0000 - val_loss: 0.1233 - val_binary_accuracy: 0.9825\n",
+      "Epoch 45/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 3.6566e-05 - binary_accuracy: 1.0000 - val_loss: 0.1243 - val_binary_accuracy: 0.9825\n",
+      "Epoch 46/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 3.6356e-05 - binary_accuracy: 1.0000 - val_loss: 0.1251 - val_binary_accuracy: 0.9825\n",
+      "Epoch 47/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 3.8864e-05 - binary_accuracy: 1.0000 - val_loss: 0.1251 - val_binary_accuracy: 0.9825\n",
+      "Epoch 48/50\n",
+      "42/42 [==============================] - 10s 239ms/step - loss: 3.6076e-05 - binary_accuracy: 1.0000 - val_loss: 0.1255 - val_binary_accuracy: 0.9825\n",
+      "Epoch 49/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 3.6211e-05 - binary_accuracy: 1.0000 - val_loss: 0.1257 - val_binary_accuracy: 0.9825\n",
+      "Epoch 50/50\n",
+      "42/42 [==============================] - 10s 240ms/step - loss: 3.4930e-05 - binary_accuracy: 1.0000 - val_loss: 0.1257 - val_binary_accuracy: 0.9825\n",
+      "18/18 [==============================] - 1s 64ms/step - loss: 0.1257 - binary_accuracy: 0.9825\n",
+      "Loss: 0.12571147084236145\n",
+      "Accuracy: 0.9825174808502197\n",
       "dict_keys(['loss', 'binary_accuracy', 'val_loss', 'val_binary_accuracy'])\n"
      ]
     },
@@ -118,19 +203,147 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "INFO:tensorflow:Assets written to: ./bert_models/1670456213.706355/assets\n"
+      "INFO:tensorflow:Assets written to: ./bert_models/1670687985.7670703/assets\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO:tensorflow:Assets written to: ./bert_models/1670456213.706355/assets\n"
+      "INFO:tensorflow:Assets written to: ./bert_models/1670687985.7670703/assets\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Message: \"Greg, can you call me back once you get this?\"\n",
+      "Likeliness of spam in percentage: 0.000008\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Congrats on your new iPhone! Click here to claim your prize...\"\n",
+      "Likeliness of spam in percentage: 0.001219\n",
+      "Vote by AI: Not Spam\n",
+      "Model failed to predict correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Really like that new photo of you\"\n",
+      "Likeliness of spam in percentage: 0.002050\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Did you hear the news today? Terrible what has happened...\"\n",
+      "Likeliness of spam in percentage: 0.000008\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Attend this free COVID webinar today: Book your session now...\"\n",
+      "Likeliness of spam in percentage: 0.999917\n",
+      "Vote by AI: Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Are you coming to the party tonight?\"\n",
+      "Likeliness of spam in percentage: 0.055407\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Your parcel has gone missing\"\n",
+      "Likeliness of spam in percentage: 0.000358\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Do not forget to bring friends!\"\n",
+      "Likeliness of spam in percentage: 0.000066\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"You have won a million dollars! Fill out your bank details here...\"\n",
+      "Likeliness of spam in percentage: 0.000235\n",
+      "Vote by AI: Not Spam\n",
+      "Model failed to predict correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Looking forward to seeing you again\"\n",
+      "Likeliness of spam in percentage: 0.484317\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"oh wow https://github.com/MGCodesandStats/tensorflow-nlp/blob/master/spam%20detection%20tensorflow%20v2.ipynb works really good on spam detection. Guess I go with that as the base model then lol :D\"\n",
+      "Likeliness of spam in percentage: 0.999941\n",
+      "Vote by AI: Spam\n",
+      "Model failed to predict correctly\n",
+      "\n",
+      "\n",
+      "Message: \"ayo\"\n",
+      "Likeliness of spam in percentage: 0.010082\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Almost all my spam is coming to my non-gmail address actually\"\n",
+      "Likeliness of spam in percentage: 0.000035\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Oh neat I think I found the sizing sweetspot for my data :D\"\n",
+      "Likeliness of spam in percentage: 0.000008\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"would never click on buttons in gmail :D always expecting there to be a bug in gmail that allows js to grab your google credentials :D XSS via email lol. I am too scared for touching spam in gmail\"\n",
+      "Likeliness of spam in percentage: 0.017660\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"back to cacophony \"\n",
+      "Likeliness of spam in percentage: 0.975112\n",
+      "Vote by AI: Spam\n",
+      "Model failed to predict correctly\n",
+      "\n",
+      "\n",
+      "Message: \"Room version 11 when\"\n",
+      "Likeliness of spam in percentage: 0.993125\n",
+      "Vote by AI: Spam\n",
+      "Model failed to predict correctly\n",
+      "\n",
+      "\n",
+      "Message: \"skip 11 and go straight to 12\"\n",
+      "Likeliness of spam in percentage: 0.100031\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"100 events should clear out any events that might be causing a request to fail lol\"\n",
+      "Likeliness of spam in percentage: 0.007106\n",
+      "Vote by AI: Not Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "Message: \"I'll help anyone interested on how to invest and earn $30k, $50k, $100k, $200k or more in just 72hours from the crypto market.But you will have to pay me my commission! when you receive your profit! if interested send me a direct message let's get started or via WhatsApp +1 (605) 953‑6801\"\n",
+      "Likeliness of spam in percentage: 0.999975\n",
+      "Vote by AI: Spam\n",
+      "Model predicted correctly\n",
+      "\n",
+      "\n",
+      "15 out of 20 are detected correctly\n",
+      "\n"
      ]
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 1000x600 with 2 Axes>"
       ]
@@ -197,21 +410,12 @@
     "batch_size = 32\n",
     "\n",
     "raw_train_ds = tf.data.Dataset.from_tensor_slices((train_examples,train_labels))\n",
-    "print(f\"raw_train_ds: {raw_train_ds}\")\n",
-    "\n",
     "train_ds = raw_train_ds.batch(batch_size).cache().prefetch(buffer_size=AUTOTUNE)\n",
-    "print(f\"train_ds: {train_ds}\")\n",
+    "\n",
     "raw_val_ds = tf.data.Dataset.from_tensor_slices((test_examples,test_labels))\n",
-    "print(f\"raw_val_ds: {raw_val_ds}\")\n",
     "val_ds = raw_val_ds.batch(batch_size).cache().prefetch(buffer_size=AUTOTUNE)\n",
     "test_ds = raw_val_ds.batch(batch_size).cache().prefetch(buffer_size=AUTOTUNE)\n",
     "\n",
-    "for text_batch, label in raw_train_ds.take(1).cache():\n",
-    "  print(f'Message: {text_batch}')\n",
-    "  print(f'Label: {label}')\n",
-    "\n",
-    "\n",
-    "\n",
     "# Load the BERT encoder and preprocessing models\n",
     "# Alternative https://tfhub.dev/google/electra_small/2\n",
     "tfhub_handle_preprocess = 'https://tfhub.dev/tensorflow/bert_en_uncased_preprocess/3'\n",
@@ -238,7 +442,7 @@
     "loss = tf.keras.losses.BinaryCrossentropy(from_logits=True)\n",
     "metrics = tf.metrics.BinaryAccuracy()\n",
     "\n",
-    "epochs = 5\n",
+    "epochs = 50\n",
     "steps_per_epoch = tf.data.experimental.cardinality(train_ds).numpy()\n",
     "num_train_steps = steps_per_epoch * epochs\n",
     "num_warmup_steps = int(0.1*num_train_steps)\n",
@@ -265,7 +469,6 @@
     "\n",
     "\n",
     "history_dict = history.history\n",
-    "print(history_dict.keys())\n",
     "\n",
     "acc = history_dict['binary_accuracy']\n",
     "val_acc = history_dict['val_binary_accuracy']\n",
@@ -299,7 +502,81 @@
     "\n",
     "classifier_model.save(saved_model_path, include_optimizer=False)\n",
     "\n",
-    "\n"
+    "def test_model(model):\n",
+    "    # Use the model to predict whether a message is spam\n",
+    "    text_messages = [\n",
+    "        \"Greg, can you call me back once you get this?\",\n",
+    "        \"Congrats on your new iPhone! Click here to claim your prize...\",\n",
+    "        \"Really like that new photo of you\",\n",
+    "        \"Did you hear the news today? Terrible what has happened...\",\n",
+    "        \"Attend this free COVID webinar today: Book your session now...\",\n",
+    "        \"Are you coming to the party tonight?\",\n",
+    "        \"Your parcel has gone missing\",\n",
+    "        \"Do not forget to bring friends!\",\n",
+    "        \"You have won a million dollars! Fill out your bank details here...\",\n",
+    "        \"Looking forward to seeing you again\",\n",
+    "        \"oh wow https://github.com/MGCodesandStats/tensorflow-nlp/blob/master/spam%20detection%20tensorflow%20v2.ipynb works really good on spam detection. Guess I go with that as the base model then lol :D\",\n",
+    "        \"ayo\",\n",
+    "        \"Almost all my spam is coming to my non-gmail address actually\",\n",
+    "        \"Oh neat I think I found the sizing sweetspot for my data :D\",\n",
+    "        \"would never click on buttons in gmail :D always expecting there to be a bug in gmail that allows js to grab your google credentials :D XSS via email lol. I am too scared for touching spam in gmail\",\n",
+    "        \"back to cacophony \",\n",
+    "        \"Room version 11 when\",\n",
+    "        \"skip 11 and go straight to 12\",\n",
+    "        \"100 events should clear out any events that might be causing a request to fail lol\",\n",
+    "        \"I'll help anyone interested on how to invest and earn $30k, $50k, $100k, $200k or more in just 72hours from the crypto market.But you will have to pay me my commission! when you receive your profit! if interested send me a direct message let's get started or via WhatsApp +1 (605) 953‑6801\",\n",
+    "    ]\n",
+    "\n",
+    "    spam_no_spam = [\n",
+    "        False,\n",
+    "        True,\n",
+    "        False,\n",
+    "        False,\n",
+    "        True,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        True,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        False,\n",
+    "        True,\n",
+    "    ]\n",
+    "\n",
+    "    # print(text_messages)\n",
+    "\n",
+    "    # Create the sequences\n",
+    "    results = tf.sigmoid(model(tf.constant(text_messages)))\n",
+    "\n",
+    "    # The closer the class is to 1, the more likely that the message is spam\n",
+    "    correct = 0\n",
+    "    expected = len(spam_no_spam)\n",
+    "    for x in range(len(text_messages)):\n",
+    "        print(f'Message: \"{text_messages[x]}\"')\n",
+    "        print(f\"Likeliness of spam in percentage: {results[x][0]:.6f}\")\n",
+    "        spam = results[x][0] >= 0.8\n",
+    "        if spam:\n",
+    "            print(\"Vote by AI: Spam\")\n",
+    "        else:\n",
+    "            print(\"Vote by AI: Not Spam\")\n",
+    "\n",
+    "        if spam_no_spam[x] != spam:\n",
+    "            print(\"Model failed to predict correctly\")\n",
+    "        else:\n",
+    "            correct = correct+1\n",
+    "            print(\"Model predicted correctly\")\n",
+    "        print(\"\\n\")\n",
+    "    print(f\"{correct} out of {expected} are detected correctly\\n\")\n",
+    "\n",
+    "\n",
+    "test_model(classifier_model)\n"
    ]
   }
  ],
diff --git a/model_v2.py b/model_v2.py
@@ -127,6 +127,7 @@ def remove_stopwords(input_text):
 def change_labels(x):
     return 1 if x == "spam" else 0
 
+
 def load_data():
     data = pd.read_csv(
         "./input/MatrixData.tsv", sep="\t", quoting=csv.QUOTE_NONE, encoding="utf-8"
@@ -257,7 +258,7 @@ def train_model(
             # tf.keras.callbacks.ModelCheckpoint(
             #    filepath=checkpoint_prefix, save_weights_only=True
             # ),
-            progress_bar
+            progress_bar,
         ],
     )
 
@@ -327,6 +328,8 @@ def test_model(model):
     classes = model.predict(np.array(text_messages))
 
     # The closer the class is to 1, the more likely that the message is spam
+    correct = 0
+    expected = len(spam_no_spam)
     for x in range(len(text_messages)):
         print(f'Message: "{text_messages[x]}"')
         print(f"Likeliness of spam in percentage: {classes[x][0]:.5f}")
@@ -338,8 +341,10 @@ def test_model(model):
         if spam_no_spam[x] != spam:
             print("Model failed to predict correctly")
         else:
+            correct = correct + 1
             print("Model predicted correctly")
         print("\n")
+    print(f"{correct} out of {expected} are detected correctly\n")
 
 
 def main():
@@ -355,19 +360,19 @@ def main():
 
     model = SpamDectionHyperModel()
     tuner = kt.Hyperband(
-       model,
-       objective="val_accuracy",
-       max_epochs=100,
-       directory="hyper_tuning",
-       project_name="spam-keras",
+        model,
+        objective="val_accuracy",
+        max_epochs=100,
+        directory="hyper_tuning",
+        project_name="spam-keras",
     )
     print("[Step 3/6] Tuning hypervalues")
     best_hps = train_hyperparamters(
-       training_sentences_final,
-       testing_sentences_final,
-       training_labels_final,
-       testing_labels_final,
-       tuner,
+        training_sentences_final,
+        testing_sentences_final,
+        training_labels_final,
+        testing_labels_final,
+        tuner,
     )
 
     print("[Step 4/6] Training model")