juneja-codebase 0.1.4__tar.gz → 2.4.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- juneja_codebase-2.4.0/LICENSE +0 -0
- juneja_codebase-2.4.0/MANIFEST.in +0 -0
- juneja_codebase-2.4.0/PKG-INFO +18 -0
- juneja_codebase-2.4.0/README.md +0 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/Denoised_Autoencoders.ipynb +316 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/Fashion_Mnist_DenseNet201_VGG19_PreTModel.ipynb +151 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/Image_Compression_Autoencoders.ipynb +300 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/boston_housing.ipynb +1798 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/cnn_fashionmnist.ipynb +248 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/posneg_imdb_ffnn..ipynb +877 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/reuters.ipynb +319 -0
- juneja_codebase-2.4.0/juneja_codebase/templates/deep_learning/rnnlstm_timeseries_imdb.ipynb +1655 -0
- juneja_codebase-2.4.0/juneja_codebase.egg-info/PKG-INFO +18 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase.egg-info/SOURCES.txt +8 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/setup.cfg +1 -1
- juneja_codebase-2.4.0/setup.py +30 -0
- juneja_codebase-0.1.4/LICENSE +0 -21
- juneja_codebase-0.1.4/MANIFEST.in +0 -3
- juneja_codebase-0.1.4/PKG-INFO +0 -75
- juneja_codebase-0.1.4/README.md +0 -57
- juneja_codebase-0.1.4/juneja_codebase.egg-info/PKG-INFO +0 -75
- juneja_codebase-0.1.4/setup.py +0 -29
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/__init__.py +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/main.py +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical10_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical10_CD.y +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical11_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical11_CD.y +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical12_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical12_CD.y +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical13_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical13_CD.y +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical1_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical2_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical3_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical4_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical5_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical6_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical7_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical8_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical9_CD.l +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/compiler_design/AnshJuneja_Practical9_CD.y +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/1_try.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/2_try.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/3_try.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/4_try.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/5_try.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/6_try.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase/templates/social_network_analysis/new.ipynb +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase.egg-info/dependency_links.txt +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase.egg-info/entry_points.txt +0 -0
- {juneja_codebase-0.1.4 → juneja_codebase-2.4.0}/juneja_codebase.egg-info/top_level.txt +0 -0
|
File without changes
|
|
File without changes
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
Metadata-Version: 2.1
|
|
2
|
+
Name: juneja_codebase
|
|
3
|
+
Version: 2.4.0
|
|
4
|
+
Summary: CLI tool to generate academic practical code files for Compiler Design, Data Structures, OS, and DBMS
|
|
5
|
+
Home-page: UNKNOWN
|
|
6
|
+
Author: AJ
|
|
7
|
+
License: UNKNOWN
|
|
8
|
+
Platform: UNKNOWN
|
|
9
|
+
Classifier: Programming Language :: Python :: 3
|
|
10
|
+
Classifier: License :: OSI Approved :: MIT License
|
|
11
|
+
Classifier: Operating System :: OS Independent
|
|
12
|
+
Classifier: Intended Audience :: Education
|
|
13
|
+
Requires-Python: >=3.6
|
|
14
|
+
Description-Content-Type: text/markdown
|
|
15
|
+
License-File: LICENSE
|
|
16
|
+
|
|
17
|
+
UNKNOWN
|
|
18
|
+
|
|
File without changes
|
|
@@ -0,0 +1,316 @@
|
|
|
1
|
+
{
|
|
2
|
+
"cells": [
|
|
3
|
+
{
|
|
4
|
+
"cell_type": "code",
|
|
5
|
+
"execution_count": null,
|
|
6
|
+
"id": "51c87197",
|
|
7
|
+
"metadata": {},
|
|
8
|
+
"outputs": [
|
|
9
|
+
{
|
|
10
|
+
"name": "stderr",
|
|
11
|
+
"output_type": "stream",
|
|
12
|
+
"text": [
|
|
13
|
+
"2025-12-08 23:32:05.261227: I external/local_xla/xla/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\n",
|
|
14
|
+
"2025-12-08 23:32:05.479630: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
|
|
15
|
+
"To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
|
|
16
|
+
"2025-12-08 23:32:07.999532: I external/local_xla/xla/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\n"
|
|
17
|
+
]
|
|
18
|
+
},
|
|
19
|
+
{
|
|
20
|
+
"name": "stdout",
|
|
21
|
+
"output_type": "stream",
|
|
22
|
+
"text": [
|
|
23
|
+
"Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
|
|
24
|
+
"\u001b[1m11490434/11490434\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 1us/step\n"
|
|
25
|
+
]
|
|
26
|
+
},
|
|
27
|
+
{
|
|
28
|
+
"name": "stderr",
|
|
29
|
+
"output_type": "stream",
|
|
30
|
+
"text": [
|
|
31
|
+
"2025-12-08 23:32:28.258002: E external/local_xla/xla/stream_executor/cuda/cuda_platform.cc:51] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)\n"
|
|
32
|
+
]
|
|
33
|
+
},
|
|
34
|
+
{
|
|
35
|
+
"data": {
|
|
36
|
+
"text/html": [
|
|
37
|
+
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional\"</span>\n",
|
|
38
|
+
"</pre>\n"
|
|
39
|
+
],
|
|
40
|
+
"text/plain": [
|
|
41
|
+
"\u001b[1mModel: \"functional\"\u001b[0m\n"
|
|
42
|
+
]
|
|
43
|
+
},
|
|
44
|
+
"metadata": {},
|
|
45
|
+
"output_type": "display_data"
|
|
46
|
+
},
|
|
47
|
+
{
|
|
48
|
+
"data": {
|
|
49
|
+
"text/html": [
|
|
50
|
+
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
|
|
51
|
+
"┃<span style=\"font-weight: bold\"> Layer (type) </span>┃<span style=\"font-weight: bold\"> Output Shape </span>┃<span style=\"font-weight: bold\"> Param # </span>┃\n",
|
|
52
|
+
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
|
|
53
|
+
"│ input_layer (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">784</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
|
|
54
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
55
|
+
"│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">100,480</span> │\n",
|
|
56
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
57
|
+
"│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">8,256</span> │\n",
|
|
58
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
59
|
+
"│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2,080</span> │\n",
|
|
60
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
61
|
+
"│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2,112</span> │\n",
|
|
62
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
63
|
+
"│ dense_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">8,320</span> │\n",
|
|
64
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
65
|
+
"│ dense_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">784</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">101,136</span> │\n",
|
|
66
|
+
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
|
|
67
|
+
"</pre>\n"
|
|
68
|
+
],
|
|
69
|
+
"text/plain": [
|
|
70
|
+
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
|
|
71
|
+
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
|
|
72
|
+
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
|
|
73
|
+
"│ input_layer (\u001b[38;5;33mInputLayer\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m784\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
|
|
74
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
75
|
+
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m100,480\u001b[0m │\n",
|
|
76
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
77
|
+
"│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m8,256\u001b[0m │\n",
|
|
78
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
79
|
+
"│ dense_2 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m2,080\u001b[0m │\n",
|
|
80
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
81
|
+
"│ dense_3 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m2,112\u001b[0m │\n",
|
|
82
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
83
|
+
"│ dense_4 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m8,320\u001b[0m │\n",
|
|
84
|
+
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
85
|
+
"│ dense_5 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m784\u001b[0m) │ \u001b[38;5;34m101,136\u001b[0m │\n",
|
|
86
|
+
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
|
|
87
|
+
]
|
|
88
|
+
},
|
|
89
|
+
"metadata": {},
|
|
90
|
+
"output_type": "display_data"
|
|
91
|
+
},
|
|
92
|
+
{
|
|
93
|
+
"data": {
|
|
94
|
+
"text/html": [
|
|
95
|
+
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">222,384</span> (868.69 KB)\n",
|
|
96
|
+
"</pre>\n"
|
|
97
|
+
],
|
|
98
|
+
"text/plain": [
|
|
99
|
+
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m222,384\u001b[0m (868.69 KB)\n"
|
|
100
|
+
]
|
|
101
|
+
},
|
|
102
|
+
"metadata": {},
|
|
103
|
+
"output_type": "display_data"
|
|
104
|
+
},
|
|
105
|
+
{
|
|
106
|
+
"data": {
|
|
107
|
+
"text/html": [
|
|
108
|
+
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">222,384</span> (868.69 KB)\n",
|
|
109
|
+
"</pre>\n"
|
|
110
|
+
],
|
|
111
|
+
"text/plain": [
|
|
112
|
+
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m222,384\u001b[0m (868.69 KB)\n"
|
|
113
|
+
]
|
|
114
|
+
},
|
|
115
|
+
"metadata": {},
|
|
116
|
+
"output_type": "display_data"
|
|
117
|
+
},
|
|
118
|
+
{
|
|
119
|
+
"data": {
|
|
120
|
+
"text/html": [
|
|
121
|
+
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
|
|
122
|
+
"</pre>\n"
|
|
123
|
+
],
|
|
124
|
+
"text/plain": [
|
|
125
|
+
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
|
|
126
|
+
]
|
|
127
|
+
},
|
|
128
|
+
"metadata": {},
|
|
129
|
+
"output_type": "display_data"
|
|
130
|
+
},
|
|
131
|
+
{
|
|
132
|
+
"name": "stdout",
|
|
133
|
+
"output_type": "stream",
|
|
134
|
+
"text": [
|
|
135
|
+
"Training the Denoising Autoencoder...\n"
|
|
136
|
+
]
|
|
137
|
+
},
|
|
138
|
+
{
|
|
139
|
+
"name": "stderr",
|
|
140
|
+
"output_type": "stream",
|
|
141
|
+
"text": [
|
|
142
|
+
"2025-12-08 23:32:28.636186: W external/local_xla/xla/tsl/framework/cpu_allocator_impl.cc:84] Allocation of 188160000 exceeds 10% of free system memory.\n",
|
|
143
|
+
"2025-12-08 23:32:28.894899: W external/local_xla/xla/tsl/framework/cpu_allocator_impl.cc:84] Allocation of 188160000 exceeds 10% of free system memory.\n"
|
|
144
|
+
]
|
|
145
|
+
},
|
|
146
|
+
{
|
|
147
|
+
"name": "stdout",
|
|
148
|
+
"output_type": "stream",
|
|
149
|
+
"text": [
|
|
150
|
+
"Epoch 1/20\n",
|
|
151
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 9ms/step - loss: 0.0689 - val_loss: 0.0470\n",
|
|
152
|
+
"Epoch 2/20\n",
|
|
153
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 8ms/step - loss: 0.0409 - val_loss: 0.0365\n",
|
|
154
|
+
"Epoch 3/20\n",
|
|
155
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 10ms/step - loss: 0.0342 - val_loss: 0.0315\n",
|
|
156
|
+
"Epoch 4/20\n",
|
|
157
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 23ms/step - loss: 0.0305 - val_loss: 0.0289\n",
|
|
158
|
+
"Epoch 5/20\n",
|
|
159
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0283 - val_loss: 0.0270\n",
|
|
160
|
+
"Epoch 6/20\n",
|
|
161
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0268 - val_loss: 0.0259\n",
|
|
162
|
+
"Epoch 7/20\n",
|
|
163
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0257 - val_loss: 0.0251\n",
|
|
164
|
+
"Epoch 8/20\n",
|
|
165
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0247 - val_loss: 0.0241\n",
|
|
166
|
+
"Epoch 9/20\n",
|
|
167
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0239 - val_loss: 0.0233\n",
|
|
168
|
+
"Epoch 10/20\n",
|
|
169
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0231 - val_loss: 0.0230\n",
|
|
170
|
+
"Epoch 11/20\n",
|
|
171
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0225 - val_loss: 0.0223\n",
|
|
172
|
+
"Epoch 12/20\n",
|
|
173
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0220 - val_loss: 0.0217\n",
|
|
174
|
+
"Epoch 13/20\n",
|
|
175
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0216 - val_loss: 0.0213\n",
|
|
176
|
+
"Epoch 14/20\n",
|
|
177
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 23ms/step - loss: 0.0211 - val_loss: 0.0211\n",
|
|
178
|
+
"Epoch 15/20\n",
|
|
179
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 20ms/step - loss: 0.0209 - val_loss: 0.0207\n",
|
|
180
|
+
"Epoch 16/20\n",
|
|
181
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 21ms/step - loss: 0.0205 - val_loss: 0.0204\n",
|
|
182
|
+
"Epoch 17/20\n",
|
|
183
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 23ms/step - loss: 0.0203 - val_loss: 0.0203\n",
|
|
184
|
+
"Epoch 18/20\n",
|
|
185
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 29ms/step - loss: 0.0201 - val_loss: 0.0200\n",
|
|
186
|
+
"Epoch 19/20\n",
|
|
187
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 24ms/step - loss: 0.0199 - val_loss: 0.0199\n",
|
|
188
|
+
"Epoch 20/20\n",
|
|
189
|
+
"\u001b[1m235/235\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 23ms/step - loss: 0.0197 - val_loss: 0.0196\n",
|
|
190
|
+
"\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 6ms/step\n"
|
|
191
|
+
]
|
|
192
|
+
},
|
|
193
|
+
{
|
|
194
|
+
"data": {
|
|
195
|
+
"image/png": "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",
|
|
196
|
+
"text/plain": [
|
|
197
|
+
"<Figure size 2000x600 with 30 Axes>"
|
|
198
|
+
]
|
|
199
|
+
},
|
|
200
|
+
"metadata": {},
|
|
201
|
+
"output_type": "display_data"
|
|
202
|
+
}
|
|
203
|
+
],
|
|
204
|
+
"source": [
|
|
205
|
+
"import tensorflow as tf\n",
|
|
206
|
+
"from tensorflow.keras.layers import Input, Dense\n",
|
|
207
|
+
"from tensorflow.keras.models import Model\n",
|
|
208
|
+
"import matplotlib.pyplot as plt\n",
|
|
209
|
+
"import numpy as np\n",
|
|
210
|
+
"\n",
|
|
211
|
+
"\n",
|
|
212
|
+
"(x_train, _), (x_test, _) = tf.keras.datasets.mnist.load_data()\n",
|
|
213
|
+
"\n",
|
|
214
|
+
"x_train = x_train.astype('float32') / 255.0\n",
|
|
215
|
+
"x_test = x_test.astype('float32') / 255.0\n",
|
|
216
|
+
"\n",
|
|
217
|
+
"# Flatten the 28x28 images into 784-dimensional vectors\n",
|
|
218
|
+
"original_dim = 784\n",
|
|
219
|
+
"x_train = x_train.reshape((len(x_train), original_dim))\n",
|
|
220
|
+
"x_test = x_test.reshape((len(x_test), original_dim))\n",
|
|
221
|
+
"\n",
|
|
222
|
+
"noise_factor = 0.5 # How much noise to add\n",
|
|
223
|
+
"\n",
|
|
224
|
+
"x_train_noisy = x_train + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape)\n",
|
|
225
|
+
"x_test_noisy = x_test + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=x_test.shape)\n",
|
|
226
|
+
"\n",
|
|
227
|
+
"x_train_noisy = np.clip(x_train_noisy, 0.0, 1.0)\n",
|
|
228
|
+
"x_test_noisy = np.clip(x_test_noisy, 0.0, 1.0)\n",
|
|
229
|
+
"\n",
|
|
230
|
+
"\n",
|
|
231
|
+
"encoding_dim = 32\n",
|
|
232
|
+
"\n",
|
|
233
|
+
"input_img = Input(shape=(original_dim,))\n",
|
|
234
|
+
"\n",
|
|
235
|
+
"encoded = Dense(128, activation='relu')(input_img)\n",
|
|
236
|
+
"encoded = Dense(64, activation='relu')(encoded)\n",
|
|
237
|
+
"encoded = Dense(encoding_dim, activation='relu')(encoded)\n",
|
|
238
|
+
"\n",
|
|
239
|
+
"decoded = Dense(64, activation='relu')(encoded)\n",
|
|
240
|
+
"decoded = Dense(128, activation='relu')(decoded)\n",
|
|
241
|
+
"decoded = Dense(original_dim, activation='sigmoid')(decoded)\n",
|
|
242
|
+
"\n",
|
|
243
|
+
"autoencoder = Model(input_img, decoded)\n",
|
|
244
|
+
"\n",
|
|
245
|
+
"autoencoder.compile(optimizer='adam', loss='mean_squared_error')\n",
|
|
246
|
+
"autoencoder.summary()\n",
|
|
247
|
+
"\n",
|
|
248
|
+
"print(\"Training the Denoising Autoencoder...\")\n",
|
|
249
|
+
"\n",
|
|
250
|
+
"autoencoder.fit(x_train_noisy, x_train,\n",
|
|
251
|
+
" epochs=20,\n",
|
|
252
|
+
" batch_size=256,\n",
|
|
253
|
+
" shuffle=True,\n",
|
|
254
|
+
" validation_data=(x_test_noisy, x_test))\n",
|
|
255
|
+
"\n",
|
|
256
|
+
"\n",
|
|
257
|
+
"decoded_imgs = autoencoder.predict(x_test_noisy)\n",
|
|
258
|
+
"\n",
|
|
259
|
+
"n = 10 # Number of digits to display\n",
|
|
260
|
+
"plt.figure(figsize=(20, 6))\n",
|
|
261
|
+
"plt.suptitle(\"Denoising Autoencoder Results\", fontsize=16)\n",
|
|
262
|
+
"\n",
|
|
263
|
+
"for i in range(n):\n",
|
|
264
|
+
" # --- Display original images (for comparison) ---\n",
|
|
265
|
+
" ax = plt.subplot(3, n, i + 1)\n",
|
|
266
|
+
" plt.imshow(x_test[i].reshape(28, 28))\n",
|
|
267
|
+
" plt.gray()\n",
|
|
268
|
+
" ax.get_xaxis().set_visible(False)\n",
|
|
269
|
+
" ax.get_yaxis().set_visible(False)\n",
|
|
270
|
+
" if i == 0:\n",
|
|
271
|
+
" ax.set_title(\"Original Clean\\nImages\", fontsize=12)\n",
|
|
272
|
+
"\n",
|
|
273
|
+
" # --- Display noisy input images ---\n",
|
|
274
|
+
" ax = plt.subplot(3, n, i + 1 + n)\n",
|
|
275
|
+
" plt.imshow(x_test_noisy[i].reshape(28, 28))\n",
|
|
276
|
+
" plt.gray()\n",
|
|
277
|
+
" ax.get_xaxis().set_visible(False)\n",
|
|
278
|
+
" ax.get_yaxis().set_visible(False)\n",
|
|
279
|
+
" if i == 0:\n",
|
|
280
|
+
" ax.set_title(\"Noisy Input\", fontsize=12)\n",
|
|
281
|
+
"\n",
|
|
282
|
+
" # --- Display reconstructed (denoised) images ---\n",
|
|
283
|
+
" ax = plt.subplot(3, n, i + 1 + 2 * n)\n",
|
|
284
|
+
" plt.imshow(decoded_imgs[i].reshape(28, 28))\n",
|
|
285
|
+
" plt.gray()\n",
|
|
286
|
+
" ax.get_xaxis().set_visible(False)\n",
|
|
287
|
+
" ax.get_yaxis().set_visible(False)\n",
|
|
288
|
+
" if i == 0:\n",
|
|
289
|
+
" ax.set_title(\"Denoised Output\", fontsize=12)\n",
|
|
290
|
+
"\n",
|
|
291
|
+
"plt.show()"
|
|
292
|
+
]
|
|
293
|
+
}
|
|
294
|
+
],
|
|
295
|
+
"metadata": {
|
|
296
|
+
"kernelspec": {
|
|
297
|
+
"display_name": ".venv (3.10.12)",
|
|
298
|
+
"language": "python",
|
|
299
|
+
"name": "python3"
|
|
300
|
+
},
|
|
301
|
+
"language_info": {
|
|
302
|
+
"codemirror_mode": {
|
|
303
|
+
"name": "ipython",
|
|
304
|
+
"version": 3
|
|
305
|
+
},
|
|
306
|
+
"file_extension": ".py",
|
|
307
|
+
"mimetype": "text/x-python",
|
|
308
|
+
"name": "python",
|
|
309
|
+
"nbconvert_exporter": "python",
|
|
310
|
+
"pygments_lexer": "ipython3",
|
|
311
|
+
"version": "3.10.12"
|
|
312
|
+
}
|
|
313
|
+
},
|
|
314
|
+
"nbformat": 4,
|
|
315
|
+
"nbformat_minor": 5
|
|
316
|
+
}
|