PyAiNetwork 0.2.2__tar.gz → 0.2.3__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.
@@ -0,0 +1,96 @@
1
+ Metadata-Version: 2.4
2
+ Name: PyAiNetwork
3
+ Version: 0.2.3
4
+ Summary: A simple Python library for creating neural networks
5
+ Author: eyes-studio
6
+ License: MIT
7
+ Requires-Python: >=3.8
8
+ Description-Content-Type: text/markdown
9
+ License-File: LICENSE
10
+ Requires-Dist: numpy
11
+ Dynamic: license-file
12
+
13
+ # PyAiNetwork
14
+
15
+ PyAiNetork - this is a lightweight library to create small to medium-sized LM's.
16
+
17
+ ---
18
+
19
+ ## Main functions
20
+
21
+ - Creating neural networks
22
+ - neural network training
23
+ - learning and counting through numpy tables
24
+
25
+ ## Installation
26
+
27
+ ```bash
28
+ pip install PyAiNetwork
29
+ ```
30
+
31
+ ---
32
+
33
+ ## examples of use normal
34
+
35
+ ```bash
36
+ from PyAiNetwork import Network
37
+
38
+ net = Network(
39
+ 4, #input
40
+ 1, #layers
41
+ 1, #neorons on layer
42
+ 4 #output
43
+ 'gelu' #activition function
44
+ )
45
+
46
+ input = [0.2,0.4,0.6,0.8]
47
+
48
+ output = net.forward(input)
49
+ print(output)
50
+
51
+ for i in range(1000):
52
+ net.train(
53
+ input, #input
54
+ input, #output
55
+ 0.01 #learing rate
56
+ )
57
+
58
+ output = net.forward(input)
59
+ print(output)
60
+ ```
61
+ ## examples of use profi
62
+
63
+ ```bash
64
+ from PyAiNetwork import ProfNetwork
65
+
66
+ layers_neorons = [1]
67
+
68
+ net = ProfNetwork(
69
+ 4, #input
70
+ layers_neorons, #layers
71
+ 4 #output
72
+ 'gelu' #activition function
73
+ 'matrix' # math type (or every)
74
+ )
75
+
76
+ input = [0.2,0.4,0.6,0.8]
77
+
78
+ output = net.forward(input)
79
+ print(output)
80
+
81
+ for i in range(1000):
82
+ net.train(
83
+ input, #input
84
+ input, #output
85
+ 0.01 #learing rate
86
+ )
87
+
88
+ output = net.forward(input)
89
+ print(output)
90
+ ```
91
+
92
+ # GitHub
93
+
94
+ https://github.com/eyes-studio/PyAiNetwork
95
+
96
+ ---
@@ -0,0 +1,96 @@
1
+ Metadata-Version: 2.4
2
+ Name: PyAiNetwork
3
+ Version: 0.2.3
4
+ Summary: A simple Python library for creating neural networks
5
+ Author: eyes-studio
6
+ License: MIT
7
+ Requires-Python: >=3.8
8
+ Description-Content-Type: text/markdown
9
+ License-File: LICENSE
10
+ Requires-Dist: numpy
11
+ Dynamic: license-file
12
+
13
+ # PyAiNetwork
14
+
15
+ PyAiNetork - this is a lightweight library to create small to medium-sized LM's.
16
+
17
+ ---
18
+
19
+ ## Main functions
20
+
21
+ - Creating neural networks
22
+ - neural network training
23
+ - learning and counting through numpy tables
24
+
25
+ ## Installation
26
+
27
+ ```bash
28
+ pip install PyAiNetwork
29
+ ```
30
+
31
+ ---
32
+
33
+ ## examples of use normal
34
+
35
+ ```bash
36
+ from PyAiNetwork import Network
37
+
38
+ net = Network(
39
+ 4, #input
40
+ 1, #layers
41
+ 1, #neorons on layer
42
+ 4 #output
43
+ 'gelu' #activition function
44
+ )
45
+
46
+ input = [0.2,0.4,0.6,0.8]
47
+
48
+ output = net.forward(input)
49
+ print(output)
50
+
51
+ for i in range(1000):
52
+ net.train(
53
+ input, #input
54
+ input, #output
55
+ 0.01 #learing rate
56
+ )
57
+
58
+ output = net.forward(input)
59
+ print(output)
60
+ ```
61
+ ## examples of use profi
62
+
63
+ ```bash
64
+ from PyAiNetwork import ProfNetwork
65
+
66
+ layers_neorons = [1]
67
+
68
+ net = ProfNetwork(
69
+ 4, #input
70
+ layers_neorons, #layers
71
+ 4 #output
72
+ 'gelu' #activition function
73
+ 'matrix' # math type (or every)
74
+ )
75
+
76
+ input = [0.2,0.4,0.6,0.8]
77
+
78
+ output = net.forward(input)
79
+ print(output)
80
+
81
+ for i in range(1000):
82
+ net.train(
83
+ input, #input
84
+ input, #output
85
+ 0.01 #learing rate
86
+ )
87
+
88
+ output = net.forward(input)
89
+ print(output)
90
+ ```
91
+
92
+ # GitHub
93
+
94
+ https://github.com/eyes-studio/PyAiNetwork
95
+
96
+ ---
@@ -0,0 +1,84 @@
1
+ # PyAiNetwork
2
+
3
+ PyAiNetork - this is a lightweight library to create small to medium-sized LM's.
4
+
5
+ ---
6
+
7
+ ## Main functions
8
+
9
+ - Creating neural networks
10
+ - neural network training
11
+ - learning and counting through numpy tables
12
+
13
+ ## Installation
14
+
15
+ ```bash
16
+ pip install PyAiNetwork
17
+ ```
18
+
19
+ ---
20
+
21
+ ## examples of use normal
22
+
23
+ ```bash
24
+ from PyAiNetwork import Network
25
+
26
+ net = Network(
27
+ 4, #input
28
+ 1, #layers
29
+ 1, #neorons on layer
30
+ 4 #output
31
+ 'gelu' #activition function
32
+ )
33
+
34
+ input = [0.2,0.4,0.6,0.8]
35
+
36
+ output = net.forward(input)
37
+ print(output)
38
+
39
+ for i in range(1000):
40
+ net.train(
41
+ input, #input
42
+ input, #output
43
+ 0.01 #learing rate
44
+ )
45
+
46
+ output = net.forward(input)
47
+ print(output)
48
+ ```
49
+ ## examples of use profi
50
+
51
+ ```bash
52
+ from PyAiNetwork import ProfNetwork
53
+
54
+ layers_neorons = [1]
55
+
56
+ net = ProfNetwork(
57
+ 4, #input
58
+ layers_neorons, #layers
59
+ 4 #output
60
+ 'gelu' #activition function
61
+ 'matrix' # math type (or every)
62
+ )
63
+
64
+ input = [0.2,0.4,0.6,0.8]
65
+
66
+ output = net.forward(input)
67
+ print(output)
68
+
69
+ for i in range(1000):
70
+ net.train(
71
+ input, #input
72
+ input, #output
73
+ 0.01 #learing rate
74
+ )
75
+
76
+ output = net.forward(input)
77
+ print(output)
78
+ ```
79
+
80
+ # GitHub
81
+
82
+ https://github.com/eyes-studio/PyAiNetwork
83
+
84
+ ---
@@ -5,7 +5,7 @@ build-backend = "setuptools.build_meta"
5
5
 
6
6
  [project]
7
7
  name = "PyAiNetwork"
8
- version = "0.2.2"
8
+ version = "0.2.3"
9
9
  description = "A simple Python library for creating neural networks"
10
10
  readme = "README.md"
11
11
  requires-python = ">=3.8"
@@ -1,211 +0,0 @@
1
- Metadata-Version: 2.4
2
- Name: PyAiNetwork
3
- Version: 0.2.2
4
- Summary: A simple Python library for creating neural networks
5
- Author: eyes-studio
6
- License: MIT
7
- Requires-Python: >=3.8
8
- Description-Content-Type: text/markdown
9
- License-File: LICENSE
10
- Requires-Dist: numpy
11
- Dynamic: license-file
12
-
13
- # PyAiNetwork
14
-
15
- A lightweight, open-source Python library for building and training neural networks with only a few lines of code.
16
-
17
- > Create AI models without writing hundreds of lines for neurons, layers, and weights.
18
-
19
- ---
20
-
21
- ## Installation
22
-
23
- ```bash
24
- pip install PyAiNetwork
25
- ```
26
-
27
- ---
28
-
29
- ## Features
30
-
31
- - 🚀 Beginner-friendly API
32
- - 🧠 Standard `Network`
33
- - ⚡ Advanced `ProfNetwork`
34
- - 🔥 GELU activation
35
- - 🔥 ReLU activation
36
- - 🔥 Sigmoid activation
37
- - 📈 Built-in training
38
- - 🎯 Custom network architectures
39
- - 🪶 Lightweight & fast
40
- - 🐍 Pure Python
41
- - 💚 Open Source
42
-
43
- ---
44
-
45
- # Quick Start
46
-
47
- ```python
48
- from PyAiNetwork import Network
49
-
50
- net = Network(
51
- 2, # Input neurons
52
- 2, # Hidden layers
53
- 4, # Neurons per hidden layer
54
- 1 # Output neurons
55
- )
56
-
57
- result = net.forward([0.5, 1.0])
58
-
59
- print(result)
60
- ```
61
-
62
- ---
63
-
64
- # Training
65
-
66
- ```python
67
- from PyAiNetwork import Network
68
-
69
- net = Network(2, 1, 4, 1)
70
-
71
- for _ in range(100):
72
- net.train(
73
- [1, 0],
74
- [1]
75
- )
76
-
77
- print(net.forward([1, 0]))
78
- ```
79
-
80
- ---
81
-
82
- # Activation Functions
83
-
84
- PyAiNetwork supports multiple activation functions.
85
-
86
- ### GELU
87
-
88
- ```python
89
- Network(..., activition="gelu")
90
- ```
91
-
92
- ### ReLU
93
-
94
- ```python
95
- Network(..., activition="relu")
96
- ```
97
-
98
- ### Sigmoid
99
-
100
- ```python
101
- Network(..., activition="sigmoid")
102
- ```
103
-
104
- ---
105
-
106
- # Network
107
-
108
- `Network` creates a neural network where every hidden layer contains the same number of neurons.
109
-
110
- ```python
111
- Network(
112
- input_neurons,
113
- hidden_layers,
114
- neurons_per_layer,
115
- output_neurons,
116
- activition="gelu"
117
- )
118
- ```
119
-
120
- Example
121
-
122
- ```python
123
- net = Network(
124
- 4,
125
- 3,
126
- 16,
127
- 2
128
- )
129
- ```
130
-
131
- Architecture
132
-
133
- ```
134
- 4 → 16 → 16 → 16 → 2
135
- ```
136
-
137
- ---
138
-
139
- # ProfNetwork
140
-
141
- `ProfNetwork` allows every hidden layer to have a different number of neurons.
142
-
143
- ```python
144
- from PyAiNetwork import ProfNetwork
145
-
146
- net = ProfNetwork(
147
- 2,
148
- [8, 16, 8],
149
- 1
150
- )
151
- ```
152
-
153
- Architecture
154
-
155
- ```
156
- 2 → 8 → 16 → 8 → 1
157
- ```
158
-
159
- Another example
160
-
161
- ```python
162
- net = ProfNetwork(
163
- 3,
164
- [32, 64, 64, 32],
165
- 5
166
- )
167
- ```
168
-
169
- ---
170
-
171
- # Which one should I use?
172
-
173
- | Class | Best for |
174
- |--------|----------|
175
- | Network | Simple projects and learning |
176
- | ProfNetwork | Custom architectures and larger AI models |
177
-
178
- ---
179
-
180
- # Roadmap
181
-
182
- Planned features for future releases
183
-
184
- - Adam Optimizer
185
- - Save / Load models
186
- - Batch training
187
- - More activation functions
188
- - More loss functions
189
- - Better performance
190
- - More neural network types
191
- - GPU support (planned)
192
-
193
- ---
194
-
195
- # About Eyes Studio
196
-
197
- PyAiNetwork is developed by **Eyes Studio**, an independent software developer focused on AI technologies, developer tools, and open-source software.
198
-
199
- ---
200
-
201
- # License
202
-
203
- MIT License
204
-
205
- Copyright © 2026 Eyes Studio
206
-
207
- ---
208
-
209
- # GitHub
210
-
211
- https://github.com/eyes-studio/PyAiNetwork
@@ -1,211 +0,0 @@
1
- Metadata-Version: 2.4
2
- Name: PyAiNetwork
3
- Version: 0.2.2
4
- Summary: A simple Python library for creating neural networks
5
- Author: eyes-studio
6
- License: MIT
7
- Requires-Python: >=3.8
8
- Description-Content-Type: text/markdown
9
- License-File: LICENSE
10
- Requires-Dist: numpy
11
- Dynamic: license-file
12
-
13
- # PyAiNetwork
14
-
15
- A lightweight, open-source Python library for building and training neural networks with only a few lines of code.
16
-
17
- > Create AI models without writing hundreds of lines for neurons, layers, and weights.
18
-
19
- ---
20
-
21
- ## Installation
22
-
23
- ```bash
24
- pip install PyAiNetwork
25
- ```
26
-
27
- ---
28
-
29
- ## Features
30
-
31
- - 🚀 Beginner-friendly API
32
- - 🧠 Standard `Network`
33
- - ⚡ Advanced `ProfNetwork`
34
- - 🔥 GELU activation
35
- - 🔥 ReLU activation
36
- - 🔥 Sigmoid activation
37
- - 📈 Built-in training
38
- - 🎯 Custom network architectures
39
- - 🪶 Lightweight & fast
40
- - 🐍 Pure Python
41
- - 💚 Open Source
42
-
43
- ---
44
-
45
- # Quick Start
46
-
47
- ```python
48
- from PyAiNetwork import Network
49
-
50
- net = Network(
51
- 2, # Input neurons
52
- 2, # Hidden layers
53
- 4, # Neurons per hidden layer
54
- 1 # Output neurons
55
- )
56
-
57
- result = net.forward([0.5, 1.0])
58
-
59
- print(result)
60
- ```
61
-
62
- ---
63
-
64
- # Training
65
-
66
- ```python
67
- from PyAiNetwork import Network
68
-
69
- net = Network(2, 1, 4, 1)
70
-
71
- for _ in range(100):
72
- net.train(
73
- [1, 0],
74
- [1]
75
- )
76
-
77
- print(net.forward([1, 0]))
78
- ```
79
-
80
- ---
81
-
82
- # Activation Functions
83
-
84
- PyAiNetwork supports multiple activation functions.
85
-
86
- ### GELU
87
-
88
- ```python
89
- Network(..., activition="gelu")
90
- ```
91
-
92
- ### ReLU
93
-
94
- ```python
95
- Network(..., activition="relu")
96
- ```
97
-
98
- ### Sigmoid
99
-
100
- ```python
101
- Network(..., activition="sigmoid")
102
- ```
103
-
104
- ---
105
-
106
- # Network
107
-
108
- `Network` creates a neural network where every hidden layer contains the same number of neurons.
109
-
110
- ```python
111
- Network(
112
- input_neurons,
113
- hidden_layers,
114
- neurons_per_layer,
115
- output_neurons,
116
- activition="gelu"
117
- )
118
- ```
119
-
120
- Example
121
-
122
- ```python
123
- net = Network(
124
- 4,
125
- 3,
126
- 16,
127
- 2
128
- )
129
- ```
130
-
131
- Architecture
132
-
133
- ```
134
- 4 → 16 → 16 → 16 → 2
135
- ```
136
-
137
- ---
138
-
139
- # ProfNetwork
140
-
141
- `ProfNetwork` allows every hidden layer to have a different number of neurons.
142
-
143
- ```python
144
- from PyAiNetwork import ProfNetwork
145
-
146
- net = ProfNetwork(
147
- 2,
148
- [8, 16, 8],
149
- 1
150
- )
151
- ```
152
-
153
- Architecture
154
-
155
- ```
156
- 2 → 8 → 16 → 8 → 1
157
- ```
158
-
159
- Another example
160
-
161
- ```python
162
- net = ProfNetwork(
163
- 3,
164
- [32, 64, 64, 32],
165
- 5
166
- )
167
- ```
168
-
169
- ---
170
-
171
- # Which one should I use?
172
-
173
- | Class | Best for |
174
- |--------|----------|
175
- | Network | Simple projects and learning |
176
- | ProfNetwork | Custom architectures and larger AI models |
177
-
178
- ---
179
-
180
- # Roadmap
181
-
182
- Planned features for future releases
183
-
184
- - Adam Optimizer
185
- - Save / Load models
186
- - Batch training
187
- - More activation functions
188
- - More loss functions
189
- - Better performance
190
- - More neural network types
191
- - GPU support (planned)
192
-
193
- ---
194
-
195
- # About Eyes Studio
196
-
197
- PyAiNetwork is developed by **Eyes Studio**, an independent software developer focused on AI technologies, developer tools, and open-source software.
198
-
199
- ---
200
-
201
- # License
202
-
203
- MIT License
204
-
205
- Copyright © 2026 Eyes Studio
206
-
207
- ---
208
-
209
- # GitHub
210
-
211
- https://github.com/eyes-studio/PyAiNetwork
@@ -1,199 +0,0 @@
1
- # PyAiNetwork
2
-
3
- A lightweight, open-source Python library for building and training neural networks with only a few lines of code.
4
-
5
- > Create AI models without writing hundreds of lines for neurons, layers, and weights.
6
-
7
- ---
8
-
9
- ## Installation
10
-
11
- ```bash
12
- pip install PyAiNetwork
13
- ```
14
-
15
- ---
16
-
17
- ## Features
18
-
19
- - 🚀 Beginner-friendly API
20
- - 🧠 Standard `Network`
21
- - ⚡ Advanced `ProfNetwork`
22
- - 🔥 GELU activation
23
- - 🔥 ReLU activation
24
- - 🔥 Sigmoid activation
25
- - 📈 Built-in training
26
- - 🎯 Custom network architectures
27
- - 🪶 Lightweight & fast
28
- - 🐍 Pure Python
29
- - 💚 Open Source
30
-
31
- ---
32
-
33
- # Quick Start
34
-
35
- ```python
36
- from PyAiNetwork import Network
37
-
38
- net = Network(
39
- 2, # Input neurons
40
- 2, # Hidden layers
41
- 4, # Neurons per hidden layer
42
- 1 # Output neurons
43
- )
44
-
45
- result = net.forward([0.5, 1.0])
46
-
47
- print(result)
48
- ```
49
-
50
- ---
51
-
52
- # Training
53
-
54
- ```python
55
- from PyAiNetwork import Network
56
-
57
- net = Network(2, 1, 4, 1)
58
-
59
- for _ in range(100):
60
- net.train(
61
- [1, 0],
62
- [1]
63
- )
64
-
65
- print(net.forward([1, 0]))
66
- ```
67
-
68
- ---
69
-
70
- # Activation Functions
71
-
72
- PyAiNetwork supports multiple activation functions.
73
-
74
- ### GELU
75
-
76
- ```python
77
- Network(..., activition="gelu")
78
- ```
79
-
80
- ### ReLU
81
-
82
- ```python
83
- Network(..., activition="relu")
84
- ```
85
-
86
- ### Sigmoid
87
-
88
- ```python
89
- Network(..., activition="sigmoid")
90
- ```
91
-
92
- ---
93
-
94
- # Network
95
-
96
- `Network` creates a neural network where every hidden layer contains the same number of neurons.
97
-
98
- ```python
99
- Network(
100
- input_neurons,
101
- hidden_layers,
102
- neurons_per_layer,
103
- output_neurons,
104
- activition="gelu"
105
- )
106
- ```
107
-
108
- Example
109
-
110
- ```python
111
- net = Network(
112
- 4,
113
- 3,
114
- 16,
115
- 2
116
- )
117
- ```
118
-
119
- Architecture
120
-
121
- ```
122
- 4 → 16 → 16 → 16 → 2
123
- ```
124
-
125
- ---
126
-
127
- # ProfNetwork
128
-
129
- `ProfNetwork` allows every hidden layer to have a different number of neurons.
130
-
131
- ```python
132
- from PyAiNetwork import ProfNetwork
133
-
134
- net = ProfNetwork(
135
- 2,
136
- [8, 16, 8],
137
- 1
138
- )
139
- ```
140
-
141
- Architecture
142
-
143
- ```
144
- 2 → 8 → 16 → 8 → 1
145
- ```
146
-
147
- Another example
148
-
149
- ```python
150
- net = ProfNetwork(
151
- 3,
152
- [32, 64, 64, 32],
153
- 5
154
- )
155
- ```
156
-
157
- ---
158
-
159
- # Which one should I use?
160
-
161
- | Class | Best for |
162
- |--------|----------|
163
- | Network | Simple projects and learning |
164
- | ProfNetwork | Custom architectures and larger AI models |
165
-
166
- ---
167
-
168
- # Roadmap
169
-
170
- Planned features for future releases
171
-
172
- - Adam Optimizer
173
- - Save / Load models
174
- - Batch training
175
- - More activation functions
176
- - More loss functions
177
- - Better performance
178
- - More neural network types
179
- - GPU support (planned)
180
-
181
- ---
182
-
183
- # About Eyes Studio
184
-
185
- PyAiNetwork is developed by **Eyes Studio**, an independent software developer focused on AI technologies, developer tools, and open-source software.
186
-
187
- ---
188
-
189
- # License
190
-
191
- MIT License
192
-
193
- Copyright © 2026 Eyes Studio
194
-
195
- ---
196
-
197
- # GitHub
198
-
199
- https://github.com/eyes-studio/PyAiNetwork
File without changes
File without changes