Hugging Face's logo

Alim2003
/
model

Keras
TF-Keras
Model card Files Community
1
Alim2003 commited on
Commit
1260757
•
1 Parent(s): 04e2681

Upload 8 files

Browse files
Files changed (9) hide show
  1. .gitattributes +1 -0
  2. fingerprint.pb +1 -0
  3. img_model.png +0 -0
  4. img_summary.png +0 -0
  5. keras_metadata.pb +10 -0
  6. model_code.ipynb +231 -0
  7. saved_model.pb +0 -0
  8. variables/variables.data-00000-of-00001 +3 -0
  9. variables/variables.index +0 -0
.gitattributes ADDED
@@ -0,0 +1 @@
 
 
1
+ variables/variables.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
fingerprint.pb ADDED
@@ -0,0 +1 @@
 
 
1
+ ���Ӽ���j��Ѣ�������ف�潤� ��������f(�̈��ۅ�2
img_model.png ADDED
img_summary.png ADDED
keras_metadata.pb ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ �3root"_tf_keras_sequential*�2{"name": "sequential", "trainable": true, "expects_training_arg": true, "dtype": "float32", "batch_input_shape": null, "must_restore_from_config": false, "preserve_input_structure_in_config": false, "autocast": false, "class_name": "Sequential", "config": {"name": "sequential", "layers": [{"class_name": "InputLayer", "config": {"batch_input_shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "dtype": "float32", "sparse": false, "ragged": false, "name": "conv2d_input"}}, {"class_name": "Conv2D", "config": {"name": "conv2d", "trainable": true, "dtype": "float32", "batch_input_shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "filters": 32, "kernel_size": {"class_name": "__tuple__", "items": [3, 3]}, "strides": {"class_name": "__tuple__", "items": [1, 1]}, "padding": "valid", "data_format": "channels_last", "dilation_rate": {"class_name": "__tuple__", "items": [1, 1]}, "groups": 1, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "MaxPooling2D", "config": {"name": "max_pooling2d", "trainable": true, "dtype": "float32", "pool_size": {"class_name": "__tuple__", "items": [2, 2]}, "padding": "valid", "strides": {"class_name": "__tuple__", "items": [2, 2]}, "data_format": "channels_last"}}, {"class_name": "Flatten", "config": {"name": "flatten", "trainable": true, "dtype": "float32", "data_format": "channels_last"}}, {"class_name": "Dense", "config": {"name": "dense", "trainable": true, "dtype": "float32", "units": 64, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}, {"class_name": "Dense", "config": {"name": "dense_1", "trainable": true, "dtype": "float32", "units": 10, "activation": "softmax", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}}, "bias_initializer": {"class_name": "Zeros", "config": {}}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}}]}, "shared_object_id": 12, "input_spec": [{"class_name": "InputSpec", "config": {"dtype": null, "shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "ndim": 4, "max_ndim": null, "min_ndim": null, "axes": {}}}], "build_input_shape": {"class_name": "TensorShape", "items": [null, 28, 28, 1]}, "is_graph_network": true, "full_save_spec": {"class_name": "__tuple__", "items": [[{"class_name": "TypeSpec", "type_spec": "tf.TensorSpec", "serialized": [{"class_name": "TensorShape", "items": [null, 28, 28, 1]}, "float32", "conv2d_input"]}], {}]}, "save_spec": {"class_name": "TypeSpec", "type_spec": "tf.TensorSpec", "serialized": [{"class_name": "TensorShape", "items": [null, 28, 28, 1]}, "float32", "conv2d_input"]}, "keras_version": "2.12.0", "backend": "tensorflow", "model_config": {"class_name": "Sequential", "config": {"name": "sequential", "layers": [{"class_name": "InputLayer", "config": {"batch_input_shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "dtype": "float32", "sparse": false, "ragged": false, "name": "conv2d_input"}, "shared_object_id": 0}, {"class_name": "Conv2D", "config": {"name": "conv2d", "trainable": true, "dtype": "float32", "batch_input_shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "filters": 32, "kernel_size": {"class_name": "__tuple__", "items": [3, 3]}, "strides": {"class_name": "__tuple__", "items": [1, 1]}, "padding": "valid", "data_format": "channels_last", "dilation_rate": {"class_name": "__tuple__", "items": [1, 1]}, "groups": 1, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}, "shared_object_id": 1}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 2}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 3}, {"class_name": "MaxPooling2D", "config": {"name": "max_pooling2d", "trainable": true, "dtype": "float32", "pool_size": {"class_name": "__tuple__", "items": [2, 2]}, "padding": "valid", "strides": {"class_name": "__tuple__", "items": [2, 2]}, "data_format": "channels_last"}, "shared_object_id": 4}, {"class_name": "Flatten", "config": {"name": "flatten", "trainable": true, "dtype": "float32", "data_format": "channels_last"}, "shared_object_id": 5}, {"class_name": "Dense", "config": {"name": "dense", "trainable": true, "dtype": "float32", "units": 64, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}, "shared_object_id": 6}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 7}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 8}, {"class_name": "Dense", "config": {"name": "dense_1", "trainable": true, "dtype": "float32", "units": 10, "activation": "softmax", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}, "shared_object_id": 9}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 10}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 11}]}}, "training_config": {"loss": "categorical_crossentropy", "metrics": [[{"class_name": "MeanMetricWrapper", "config": {"name": "accuracy", "dtype": "float32", "fn": "categorical_accuracy"}, "shared_object_id": 14}]], "weighted_metrics": null, "loss_weights": null, "optimizer_config": {"class_name": "Custom>Adam", "config": {"name": "Adam", "weight_decay": null, "clipnorm": null, "global_clipnorm": null, "clipvalue": null, "use_ema": false, "ema_momentum": 0.99, "ema_overwrite_frequency": null, "jit_compile": false, "is_legacy_optimizer": false, "learning_rate": 0.0010000000474974513, "beta_1": 0.9, "beta_2": 0.999, "epsilon": 1e-07, "amsgrad": false}}}}2
3
+ � root.layer_with_weights-0"_tf_keras_layer*�
4
+ {"name": "conv2d", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "stateful": false, "must_restore_from_config": false, "preserve_input_structure_in_config": false, "autocast": true, "class_name": "Conv2D", "config": {"name": "conv2d", "trainable": true, "dtype": "float32", "batch_input_shape": {"class_name": "__tuple__", "items": [null, 28, 28, 1]}, "filters": 32, "kernel_size": {"class_name": "__tuple__", "items": [3, 3]}, "strides": {"class_name": "__tuple__", "items": [1, 1]}, "padding": "valid", "data_format": "channels_last", "dilation_rate": {"class_name": "__tuple__", "items": [1, 1]}, "groups": 1, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}, "shared_object_id": 1}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 2}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 3, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 4, "axes": {"-1": 1}}, "shared_object_id": 15}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 28, 28, 1]}}2
5
+ � root.layer-1"_tf_keras_layer*�{"name": "max_pooling2d", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "preserve_input_structure_in_config": false, "autocast": true, "class_name": "MaxPooling2D", "config": {"name": "max_pooling2d", "trainable": true, "dtype": "float32", "pool_size": {"class_name": "__tuple__", "items": [2, 2]}, "padding": "valid", "strides": {"class_name": "__tuple__", "items": [2, 2]}, "data_format": "channels_last"}, "shared_object_id": 4, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": 4, "max_ndim": null, "min_ndim": null, "axes": {}}, "shared_object_id": 16}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 26, 26, 32]}}2
6
+ � root.layer-2"_tf_keras_layer*�{"name": "flatten", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "preserve_input_structure_in_config": false, "autocast": true, "class_name": "Flatten", "config": {"name": "flatten", "trainable": true, "dtype": "float32", "data_format": "channels_last"}, "shared_object_id": 5, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 1, "axes": {}}, "shared_object_id": 17}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 13, 13, 32]}}2
7
+ �root.layer_with_weights-1"_tf_keras_layer*�{"name": "dense", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "preserve_input_structure_in_config": false, "autocast": true, "class_name": "Dense", "config": {"name": "dense", "trainable": true, "dtype": "float32", "units": 64, "activation": "relu", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}, "shared_object_id": 6}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 7}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 8, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 5408}}, "shared_object_id": 18}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 5408]}}2
8
+ �root.layer_with_weights-2"_tf_keras_layer*�{"name": "dense_1", "trainable": true, "expects_training_arg": false, "dtype": "float32", "batch_input_shape": null, "stateful": false, "must_restore_from_config": false, "preserve_input_structure_in_config": false, "autocast": true, "class_name": "Dense", "config": {"name": "dense_1", "trainable": true, "dtype": "float32", "units": 10, "activation": "softmax", "use_bias": true, "kernel_initializer": {"class_name": "GlorotUniform", "config": {"seed": null}, "shared_object_id": 9}, "bias_initializer": {"class_name": "Zeros", "config": {}, "shared_object_id": 10}, "kernel_regularizer": null, "bias_regularizer": null, "activity_regularizer": null, "kernel_constraint": null, "bias_constraint": null}, "shared_object_id": 11, "input_spec": {"class_name": "InputSpec", "config": {"dtype": null, "shape": null, "ndim": null, "max_ndim": null, "min_ndim": 2, "axes": {"-1": 64}}, "shared_object_id": 19}, "build_input_shape": {"class_name": "TensorShape", "items": [null, 64]}}2
9
+ �lroot.keras_api.metrics.0"_tf_keras_metric*�{"class_name": "Mean", "name": "loss", "dtype": "float32", "config": {"name": "loss", "dtype": "float32"}, "shared_object_id": 20}2
10
+ �mroot.keras_api.metrics.1"_tf_keras_metric*�{"class_name": "MeanMetricWrapper", "name": "accuracy", "dtype": "float32", "config": {"name": "accuracy", "dtype": "float32", "fn": "categorical_accuracy"}, "shared_object_id": 14}2
model_code.ipynb ADDED
@@ -0,0 +1,231 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "nbformat": 4,
3
+ "nbformat_minor": 0,
4
+ "metadata": {
5
+ "colab": {
6
+ "provenance": []
7
+ },
8
+ "kernelspec": {
9
+ "name": "python3",
10
+ "display_name": "Python 3"
11
+ },
12
+ "language_info": {
13
+ "name": "python"
14
+ }
15
+ },
16
+ "cells": [
17
+ {
18
+ "cell_type": "code",
19
+ "execution_count": null,
20
+ "metadata": {
21
+ "id": "yFWJxpqH1Z2a"
22
+ },
23
+ "outputs": [],
24
+ "source": [
25
+ "import numpy as np\n",
26
+ "import tensorflow as tf\n",
27
+ "import matplotlib.pyplot as plt\n",
28
+ "from tensorflow.keras.datasets import mnist\n",
29
+ "from tensorflow import keras\n",
30
+ "from tensorflow.keras.models import Sequential\n",
31
+ "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense\n",
32
+ "\n",
33
+ "(x_train, y_train), (x_test, y_test) = mnist.load_data()"
34
+ ]
35
+ },
36
+ {
37
+ "cell_type": "code",
38
+ "source": [
39
+ "x_train = x_train / 255\n",
40
+ "x_test = x_test / 255\n",
41
+ "\n",
42
+ "y_train = tf.keras.utils.to_categorical(y_train, num_classes=10)\n",
43
+ "y_test = tf.keras.utils.to_categorical(y_test, num_classes=10)"
44
+ ],
45
+ "metadata": {
46
+ "id": "AqllR5uK1ylh"
47
+ },
48
+ "execution_count": null,
49
+ "outputs": []
50
+ },
51
+ {
52
+ "cell_type": "code",
53
+ "source": [
54
+ "model = Sequential()\n",
55
+ "model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)))\n",
56
+ "model.add(MaxPooling2D((2, 2)))\n",
57
+ "model.add(Flatten())\n",
58
+ "model.add(Dense(64, activation='relu'))\n",
59
+ "model.add(Dense(10, activation='softmax'))\n",
60
+ "model.summary()\n",
61
+ "\n",
62
+ "train_size = x_train.shape[0]\n",
63
+ "val_size = int(train_size * 0.1)\n",
64
+ "test_size = x_test.shape[0]\n",
65
+ "print(\"Размер тренировочного датасета:\", train_size)\n",
66
+ "print(\"Размер валидационного датасета:\", val_size)\n",
67
+ "print(\"Размер тестового датасета:\", test_size)\n",
68
+ "tf.keras.utils.plot_model(model, show_shapes= True, show_layer_names= True, show_layer_activations= True)"
69
+ ],
70
+ "metadata": {
71
+ "colab": {
72
+ "base_uri": "https://localhost:8080/",
73
+ "height": 1000
74
+ },
75
+ "id": "VV1S-bKl2DEx",
76
+ "outputId": "ec7a2178-a5e2-49df-c812-1363132c58d0"
77
+ },
78
+ "execution_count": null,
79
+ "outputs": [
80
+ {
81
+ "output_type": "stream",
82
+ "name": "stdout",
83
+ "text": [
84
+ "Model: \"sequential\"\n",
85
+ "_________________________________________________________________\n",
86
+ " Layer (type) Output Shape Param # \n",
87
+ "=================================================================\n",
88
+ " conv2d (Conv2D) (None, 26, 26, 32) 320 \n",
89
+ " \n",
90
+ " max_pooling2d (MaxPooling2D (None, 13, 13, 32) 0 \n",
91
+ " ) \n",
92
+ " \n",
93
+ " flatten (Flatten) (None, 5408) 0 \n",
94
+ " \n",
95
+ " dense (Dense) (None, 64) 346176 \n",
96
+ " \n",
97
+ " dense_1 (Dense) (None, 10) 650 \n",
98
+ " \n",
99
+ "=================================================================\n",
100
+ "Total params: 347,146\n",
101
+ "Trainable params: 347,146\n",
102
+ "Non-trainable params: 0\n",
103
+ "_________________________________________________________________\n",
104
+ "Размер тренировочного датасета: 60000\n",
105
+ "Размер валидационного датасета: 6000\n",
106
+ "Размер тестового датасета: 10000\n"
107
+ ]
108
+ },
109
+ {
110
+ "output_type": "execute_result",
111
+ "data": {
112
+ "image/png": "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\n",
113
+ "text/plain": [
114
+ "<IPython.core.display.Image object>"
115
+ ]
116
+ },
117
+ "metadata": {},
118
+ "execution_count": 9
119
+ }
120
+ ]
121
+ },
122
+ {
123
+ "cell_type": "code",
124
+ "source": [
125
+ "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
126
+ "model.fit(x_train, y_train, batch_size=32, epochs=10, validation_data=(x_test, y_test))\n"
127
+ ],
128
+ "metadata": {
129
+ "colab": {
130
+ "base_uri": "https://localhost:8080/"
131
+ },
132
+ "id": "YhSYe94J2oPH",
133
+ "outputId": "98b973f8-7f63-4550-f463-70696bb5ea7c"
134
+ },
135
+ "execution_count": null,
136
+ "outputs": [
137
+ {
138
+ "output_type": "stream",
139
+ "name": "stdout",
140
+ "text": [
141
+ "Epoch 1/10\n",
142
+ "1875/1875 [==============================] - 40s 21ms/step - loss: 0.1645 - accuracy: 0.9517 - val_loss: 0.0702 - val_accuracy: 0.9779\n",
143
+ "Epoch 2/10\n",
144
+ "1875/1875 [==============================] - 39s 21ms/step - loss: 0.0583 - accuracy: 0.9819 - val_loss: 0.0495 - val_accuracy: 0.9829\n",
145
+ "Epoch 3/10\n",
146
+ "1875/1875 [==============================] - 39s 21ms/step - loss: 0.0384 - accuracy: 0.9880 - val_loss: 0.0414 - val_accuracy: 0.9871\n",
147
+ "Epoch 4/10\n",
148
+ "1875/1875 [==============================] - 36s 19ms/step - loss: 0.0272 - accuracy: 0.9914 - val_loss: 0.0479 - val_accuracy: 0.9853\n",
149
+ "Epoch 5/10\n",
150
+ "1875/1875 [==============================] - 37s 19ms/step - loss: 0.0197 - accuracy: 0.9940 - val_loss: 0.0379 - val_accuracy: 0.9883\n",
151
+ "Epoch 6/10\n",
152
+ "1875/1875 [==============================] - 35s 18ms/step - loss: 0.0138 - accuracy: 0.9957 - val_loss: 0.0495 - val_accuracy: 0.9855\n",
153
+ "Epoch 7/10\n",
154
+ "1875/1875 [==============================] - 37s 20ms/step - loss: 0.0097 - accuracy: 0.9968 - val_loss: 0.0431 - val_accuracy: 0.9872\n",
155
+ "Epoch 8/10\n",
156
+ "1875/1875 [==============================] - 37s 20ms/step - loss: 0.0079 - accuracy: 0.9973 - val_loss: 0.0488 - val_accuracy: 0.9862\n",
157
+ "Epoch 9/10\n",
158
+ "1875/1875 [==============================] - 36s 19ms/step - loss: 0.0058 - accuracy: 0.9981 - val_loss: 0.0622 - val_accuracy: 0.9845\n",
159
+ "Epoch 10/10\n",
160
+ "1875/1875 [==============================] - 35s 19ms/step - loss: 0.0056 - accuracy: 0.9982 - val_loss: 0.0569 - val_accuracy: 0.9863\n"
161
+ ]
162
+ },
163
+ {
164
+ "output_type": "execute_result",
165
+ "data": {
166
+ "text/plain": [
167
+ "<keras.callbacks.History at 0x7fec9bc82080>"
168
+ ]
169
+ },
170
+ "metadata": {},
171
+ "execution_count": 10
172
+ }
173
+ ]
174
+ },
175
+ {
176
+ "cell_type": "code",
177
+ "source": [
178
+ "loss, accuracy = model.evaluate(x_test, y_test)\n",
179
+ "print(f'Loss: {loss}, Accuracy: {accuracy}')"
180
+ ],
181
+ "metadata": {
182
+ "colab": {
183
+ "base_uri": "https://localhost:8080/"
184
+ },
185
+ "id": "gbDwrnOx23r3",
186
+ "outputId": "55c5d1b8-b9ea-42a0-a467-6946634a97bb"
187
+ },
188
+ "execution_count": null,
189
+ "outputs": [
190
+ {
191
+ "output_type": "stream",
192
+ "name": "stdout",
193
+ "text": [
194
+ "313/313 [==============================] - 2s 6ms/step - loss: 0.0569 - accuracy: 0.9863\n",
195
+ "Loss: 0.05686398223042488, Accuracy: 0.986299991607666\n"
196
+ ]
197
+ }
198
+ ]
199
+ },
200
+ {
201
+ "cell_type": "code",
202
+ "source": [
203
+ "index = np.random.randint(len(x_test))\n",
204
+ "image = x_test[index]\n",
205
+ "image = np.expand_dims(image, axis=0)\n",
206
+ "prediction = model.predict(image)\n",
207
+ "predicted = np.argmax(prediction)\n",
208
+ "remainder = predicted % 2\n",
209
+ "print(f'Predicted Digit: {predicted}, Remainder: {remainder}')"
210
+ ],
211
+ "metadata": {
212
+ "colab": {
213
+ "base_uri": "https://localhost:8080/"
214
+ },
215
+ "id": "9iEa_hL729T9",
216
+ "outputId": "1ab179be-5445-4a63-a963-a95ff0440ed4"
217
+ },
218
+ "execution_count": null,
219
+ "outputs": [
220
+ {
221
+ "output_type": "stream",
222
+ "name": "stdout",
223
+ "text": [
224
+ "1/1 [==============================] - 0s 102ms/step\n",
225
+ "Predicted Digit: 2, Remainder: 0\n"
226
+ ]
227
+ }
228
+ ]
229
+ }
230
+ ]
231
+ }
saved_model.pb ADDED
Binary file (117 kB). View file
 
variables/variables.data-00000-of-00001 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6d05c659a03b5306a7ddf28bc3cb65495d1f61582b3b3a49a01f7e42e2a723d
3
+ size 4172269
variables/variables.index ADDED
Binary file (1.57 kB). View file