dynamic-learning-model 4.0.1__tar.gz → 4.0.3__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.
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/LICENSE +1 -1
- dynamic_learning_model-4.0.3/PKG-INFO +118 -0
- dynamic_learning_model-4.0.3/README.md +100 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dlm/DLM.py +3 -0
- dynamic_learning_model-4.0.3/dynamic_learning_model.egg-info/PKG-INFO +118 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dynamic_learning_model.egg-info/SOURCES.txt +0 -1
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/pyproject.toml +1 -1
- dynamic_learning_model-4.0.1/PKG-INFO +0 -87
- dynamic_learning_model-4.0.1/README.md +0 -64
- dynamic_learning_model-4.0.1/dynamic_learning_model.egg-info/PKG-INFO +0 -87
- dynamic_learning_model-4.0.1/setup.py +0 -32
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dlm/DLM_Compute_Model.py +0 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dlm/DLM_Memory_Model.py +0 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dlm/__init__.py +0 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dynamic_learning_model.egg-info/dependency_links.txt +0 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dynamic_learning_model.egg-info/requires.txt +0 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/dynamic_learning_model.egg-info/top_level.txt +0 -0
- {dynamic_learning_model-4.0.1 → dynamic_learning_model-4.0.3}/setup.cfg +0 -0
|
@@ -0,0 +1,118 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: dynamic-learning-model
|
|
3
|
+
Version: 4.0.3
|
|
4
|
+
Summary: A Dynamic-Learning Model (DLM) chatbot with memory and compute reasoning modes.
|
|
5
|
+
Author-email: Vignesh Thondikulam <vignesh.tho2006@gmail.com>
|
|
6
|
+
License: MIT
|
|
7
|
+
Project-URL: Homepage, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
8
|
+
Project-URL: Repository, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
9
|
+
Requires-Python: >=3.12
|
|
10
|
+
Description-Content-Type: text/markdown
|
|
11
|
+
License-File: LICENSE
|
|
12
|
+
Requires-Dist: spacy>=3.7
|
|
13
|
+
Requires-Dist: transformers>=4.30
|
|
14
|
+
Requires-Dist: torch>=2.0
|
|
15
|
+
Requires-Dist: better-profanity>=0.7
|
|
16
|
+
Requires-Dist: nltk>=3.8
|
|
17
|
+
Dynamic: license-file
|
|
18
|
+
|
|
19
|
+
<table>
|
|
20
|
+
<tr>
|
|
21
|
+
<td><img src="https://ik.imagekit.io/cqhzoyggfm/DLM%20Logo.png?updatedAt=1759635222204" width="90"></td>
|
|
22
|
+
<td><h1>Dynamic Learning Model</h1></td>
|
|
23
|
+
</tr>
|
|
24
|
+
</table>
|
|
25
|
+
|
|
26
|
+
[](https://pypi.org/project/dynamic-learning-model/)
|
|
27
|
+
[](https://pypi.org/project/dynamic-learning-model/)
|
|
28
|
+
[](LICENSE)
|
|
29
|
+
|
|
30
|
+
## Overview
|
|
31
|
+
|
|
32
|
+
The Dynamic Learning Model (DLM) is a hybrid, domain-specific AI system designed to learn, adapt, and respond intelligently to user queries. It combines natural language understanding with structured reasoning, continually improving as it is trained.
|
|
33
|
+
|
|
34
|
+
**Key capabilities include:**
|
|
35
|
+
|
|
36
|
+
- **FAQ Handling** - Learns and responds to frequently asked questions based on the knowledge it has been trained on.
|
|
37
|
+
- **Chain-of-Thought (CoT) Reasoning** - Performs clear, step-by-step logic to solve non-ambiguous arithmetic, geometric, and unit conversion problems.
|
|
38
|
+
- **Custom Knowledge Integration** - DLM is fully extensible. You can initialize it with an empty SQL database and train it with your domain-specific knowledge.
|
|
39
|
+
|
|
40
|
+
Whether you're building a student support bot, a domain-specific assistant, or a computation system, DLM offers a flexible foundation to power your intelligent applications.
|
|
41
|
+
|
|
42
|
+
## Table of Contents
|
|
43
|
+
|
|
44
|
+
- [Installation](#installation)
|
|
45
|
+
- [Required Parameters](#required-parameters)
|
|
46
|
+
- [Usage](#usage)
|
|
47
|
+
- [Important Notices](#important-notices)
|
|
48
|
+
- [License](#license)
|
|
49
|
+
- [Disclaimer](#disclaimer)
|
|
50
|
+
|
|
51
|
+
## Installation
|
|
52
|
+
|
|
53
|
+
```bash
|
|
54
|
+
pip install dynamic-learning-model
|
|
55
|
+
```
|
|
56
|
+
|
|
57
|
+
> **Requirements:** Python 3.12.0 or higher is required to use this bot in your program. All required dependencies are installed automatically with the package.
|
|
58
|
+
|
|
59
|
+
## Required Parameters
|
|
60
|
+
|
|
61
|
+
The constructor requires passing in two parameters:
|
|
62
|
+
|
|
63
|
+
1. **Bot Mode**
|
|
64
|
+
- `"learn"` - Enables training using the memory model. The bot can be updated with new information.
|
|
65
|
+
- `"apply"` - The bot automatically switches between its "compute" and "memory" models depending on the query asked.
|
|
66
|
+
2. **Database Path** - An empty (or existing) SQLite database file used for training and as the bot's memory model.
|
|
67
|
+
|
|
68
|
+
The `ask()` method also requires two parameters:
|
|
69
|
+
|
|
70
|
+
1. **Query** - e.g., `"What is the definition of FAFSA?"`
|
|
71
|
+
2. **Display Thought** - `True` to display the bot's Chain-of-Thought reasoning, or `False` to suppress it.
|
|
72
|
+
|
|
73
|
+
## Usage
|
|
74
|
+
|
|
75
|
+
**`"learn"` mode** (training queries):
|
|
76
|
+
|
|
77
|
+
```python
|
|
78
|
+
from dlm import DLM
|
|
79
|
+
|
|
80
|
+
training_bot = DLM("learn", "college_knowledge.db")
|
|
81
|
+
training_bot.ask("What is FAFSA in college?", True)
|
|
82
|
+
```
|
|
83
|
+
|
|
84
|
+
**`"apply"` mode** (deployment / production use after training):
|
|
85
|
+
|
|
86
|
+
```python
|
|
87
|
+
from dlm import DLM
|
|
88
|
+
|
|
89
|
+
commercial_bot = DLM("apply", "college_knowledge.db")
|
|
90
|
+
commercial_bot.ask("What is the difference between FAFSA and CADAA in California?", False)
|
|
91
|
+
|
|
92
|
+
# or
|
|
93
|
+
|
|
94
|
+
commercial_bot.ask("Tell me the result for the following: 5 * 5 * 5 + 5 / 5", True)
|
|
95
|
+
```
|
|
96
|
+
|
|
97
|
+
### High-Level Pipeline
|
|
98
|
+
|
|
99
|
+
|
|
100
|
+
|
|
101
|
+
## Important Notices
|
|
102
|
+
|
|
103
|
+
1. **Training data quality matters.** DLM's accuracy in `"learn"` mode depends entirely on the consistency and clarity of the question/answer pairs it's trained with. Inconsistent category labeling or vague phrasing during training can produce inaccurate or corrupted responses later.
|
|
104
|
+
2. **Database files are local and untracked.** DLM stores all trained knowledge in the SQLite file you provide. Back up this file regularly - there is no built-in cloud sync, versioning, or recovery mechanism.
|
|
105
|
+
3. **Model loading behavior.** Underlying NLP and transformer models (spaCy, HuggingFace) are lazy-loaded and shared across instances. The first call in a session may take longer due to model loading; subsequent calls are significantly faster.
|
|
106
|
+
4. **Compute mode limitations.** Chain-of-Thought computation is designed for clear, non-ambiguous arithmetic, geometric, and unit-conversion problems. Ambiguous or multi-interpretation queries may produce incorrect results - always verify outputs for critical use cases.
|
|
107
|
+
|
|
108
|
+
## License
|
|
109
|
+
|
|
110
|
+
This project is licensed under the **MIT License** - see the [LICENSE](LICENSE) file for details.
|
|
111
|
+
|
|
112
|
+
## Disclaimer
|
|
113
|
+
|
|
114
|
+
Dynamic Learning Model (DLM) is provided **"as-is"**, without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement. In no event shall the author be liable for any claim, damages, or other liability arising from the use of this software.
|
|
115
|
+
|
|
116
|
+
DLM **may produce inaccurate, incomplete, or unexpected responses**, particularly for ambiguous queries or insufficiently trained knowledge bases. **Do not rely on DLM's output for decisions involving safety, legal, medical, or financial consequences without independent verification.**
|
|
117
|
+
|
|
118
|
+
All data provided to DLM (training queries, database contents) is processed and stored **locally** on the host machine. DLM does not transmit user data externally, except for any underlying third-party model downloads (e.g. HuggingFace, spaCy) required on first run, which are subject to those providers' own terms.
|
|
@@ -0,0 +1,100 @@
|
|
|
1
|
+
<table>
|
|
2
|
+
<tr>
|
|
3
|
+
<td><img src="https://ik.imagekit.io/cqhzoyggfm/DLM%20Logo.png?updatedAt=1759635222204" width="90"></td>
|
|
4
|
+
<td><h1>Dynamic Learning Model</h1></td>
|
|
5
|
+
</tr>
|
|
6
|
+
</table>
|
|
7
|
+
|
|
8
|
+
[](https://pypi.org/project/dynamic-learning-model/)
|
|
9
|
+
[](https://pypi.org/project/dynamic-learning-model/)
|
|
10
|
+
[](LICENSE)
|
|
11
|
+
|
|
12
|
+
## Overview
|
|
13
|
+
|
|
14
|
+
The Dynamic Learning Model (DLM) is a hybrid, domain-specific AI system designed to learn, adapt, and respond intelligently to user queries. It combines natural language understanding with structured reasoning, continually improving as it is trained.
|
|
15
|
+
|
|
16
|
+
**Key capabilities include:**
|
|
17
|
+
|
|
18
|
+
- **FAQ Handling** - Learns and responds to frequently asked questions based on the knowledge it has been trained on.
|
|
19
|
+
- **Chain-of-Thought (CoT) Reasoning** - Performs clear, step-by-step logic to solve non-ambiguous arithmetic, geometric, and unit conversion problems.
|
|
20
|
+
- **Custom Knowledge Integration** - DLM is fully extensible. You can initialize it with an empty SQL database and train it with your domain-specific knowledge.
|
|
21
|
+
|
|
22
|
+
Whether you're building a student support bot, a domain-specific assistant, or a computation system, DLM offers a flexible foundation to power your intelligent applications.
|
|
23
|
+
|
|
24
|
+
## Table of Contents
|
|
25
|
+
|
|
26
|
+
- [Installation](#installation)
|
|
27
|
+
- [Required Parameters](#required-parameters)
|
|
28
|
+
- [Usage](#usage)
|
|
29
|
+
- [Important Notices](#important-notices)
|
|
30
|
+
- [License](#license)
|
|
31
|
+
- [Disclaimer](#disclaimer)
|
|
32
|
+
|
|
33
|
+
## Installation
|
|
34
|
+
|
|
35
|
+
```bash
|
|
36
|
+
pip install dynamic-learning-model
|
|
37
|
+
```
|
|
38
|
+
|
|
39
|
+
> **Requirements:** Python 3.12.0 or higher is required to use this bot in your program. All required dependencies are installed automatically with the package.
|
|
40
|
+
|
|
41
|
+
## Required Parameters
|
|
42
|
+
|
|
43
|
+
The constructor requires passing in two parameters:
|
|
44
|
+
|
|
45
|
+
1. **Bot Mode**
|
|
46
|
+
- `"learn"` - Enables training using the memory model. The bot can be updated with new information.
|
|
47
|
+
- `"apply"` - The bot automatically switches between its "compute" and "memory" models depending on the query asked.
|
|
48
|
+
2. **Database Path** - An empty (or existing) SQLite database file used for training and as the bot's memory model.
|
|
49
|
+
|
|
50
|
+
The `ask()` method also requires two parameters:
|
|
51
|
+
|
|
52
|
+
1. **Query** - e.g., `"What is the definition of FAFSA?"`
|
|
53
|
+
2. **Display Thought** - `True` to display the bot's Chain-of-Thought reasoning, or `False` to suppress it.
|
|
54
|
+
|
|
55
|
+
## Usage
|
|
56
|
+
|
|
57
|
+
**`"learn"` mode** (training queries):
|
|
58
|
+
|
|
59
|
+
```python
|
|
60
|
+
from dlm import DLM
|
|
61
|
+
|
|
62
|
+
training_bot = DLM("learn", "college_knowledge.db")
|
|
63
|
+
training_bot.ask("What is FAFSA in college?", True)
|
|
64
|
+
```
|
|
65
|
+
|
|
66
|
+
**`"apply"` mode** (deployment / production use after training):
|
|
67
|
+
|
|
68
|
+
```python
|
|
69
|
+
from dlm import DLM
|
|
70
|
+
|
|
71
|
+
commercial_bot = DLM("apply", "college_knowledge.db")
|
|
72
|
+
commercial_bot.ask("What is the difference between FAFSA and CADAA in California?", False)
|
|
73
|
+
|
|
74
|
+
# or
|
|
75
|
+
|
|
76
|
+
commercial_bot.ask("Tell me the result for the following: 5 * 5 * 5 + 5 / 5", True)
|
|
77
|
+
```
|
|
78
|
+
|
|
79
|
+
### High-Level Pipeline
|
|
80
|
+
|
|
81
|
+
|
|
82
|
+
|
|
83
|
+
## Important Notices
|
|
84
|
+
|
|
85
|
+
1. **Training data quality matters.** DLM's accuracy in `"learn"` mode depends entirely on the consistency and clarity of the question/answer pairs it's trained with. Inconsistent category labeling or vague phrasing during training can produce inaccurate or corrupted responses later.
|
|
86
|
+
2. **Database files are local and untracked.** DLM stores all trained knowledge in the SQLite file you provide. Back up this file regularly - there is no built-in cloud sync, versioning, or recovery mechanism.
|
|
87
|
+
3. **Model loading behavior.** Underlying NLP and transformer models (spaCy, HuggingFace) are lazy-loaded and shared across instances. The first call in a session may take longer due to model loading; subsequent calls are significantly faster.
|
|
88
|
+
4. **Compute mode limitations.** Chain-of-Thought computation is designed for clear, non-ambiguous arithmetic, geometric, and unit-conversion problems. Ambiguous or multi-interpretation queries may produce incorrect results - always verify outputs for critical use cases.
|
|
89
|
+
|
|
90
|
+
## License
|
|
91
|
+
|
|
92
|
+
This project is licensed under the **MIT License** - see the [LICENSE](LICENSE) file for details.
|
|
93
|
+
|
|
94
|
+
## Disclaimer
|
|
95
|
+
|
|
96
|
+
Dynamic Learning Model (DLM) is provided **"as-is"**, without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement. In no event shall the author be liable for any claim, damages, or other liability arising from the use of this software.
|
|
97
|
+
|
|
98
|
+
DLM **may produce inaccurate, incomplete, or unexpected responses**, particularly for ambiguous queries or insufficiently trained knowledge bases. **Do not rely on DLM's output for decisions involving safety, legal, medical, or financial consequences without independent verification.**
|
|
99
|
+
|
|
100
|
+
All data provided to DLM (training queries, database contents) is processed and stored **locally** on the host machine. DLM does not transmit user data externally, except for any underlying third-party model downloads (e.g. HuggingFace, spaCy) required on first run, which are subject to those providers' own terms.
|
|
@@ -4,6 +4,9 @@ import random
|
|
|
4
4
|
import spacy
|
|
5
5
|
import time
|
|
6
6
|
import sqlite3
|
|
7
|
+
from transformers.utils import logging as hf_logging
|
|
8
|
+
hf_logging.disable_progress_bar()
|
|
9
|
+
hf_logging.set_verbosity_error()
|
|
7
10
|
from .DLM_Compute_Model import perform_advanced_CoT
|
|
8
11
|
from .DLM_Memory_Model import get_category
|
|
9
12
|
from .DLM_Memory_Model import get_specific_question
|
|
@@ -0,0 +1,118 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: dynamic-learning-model
|
|
3
|
+
Version: 4.0.3
|
|
4
|
+
Summary: A Dynamic-Learning Model (DLM) chatbot with memory and compute reasoning modes.
|
|
5
|
+
Author-email: Vignesh Thondikulam <vignesh.tho2006@gmail.com>
|
|
6
|
+
License: MIT
|
|
7
|
+
Project-URL: Homepage, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
8
|
+
Project-URL: Repository, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
9
|
+
Requires-Python: >=3.12
|
|
10
|
+
Description-Content-Type: text/markdown
|
|
11
|
+
License-File: LICENSE
|
|
12
|
+
Requires-Dist: spacy>=3.7
|
|
13
|
+
Requires-Dist: transformers>=4.30
|
|
14
|
+
Requires-Dist: torch>=2.0
|
|
15
|
+
Requires-Dist: better-profanity>=0.7
|
|
16
|
+
Requires-Dist: nltk>=3.8
|
|
17
|
+
Dynamic: license-file
|
|
18
|
+
|
|
19
|
+
<table>
|
|
20
|
+
<tr>
|
|
21
|
+
<td><img src="https://ik.imagekit.io/cqhzoyggfm/DLM%20Logo.png?updatedAt=1759635222204" width="90"></td>
|
|
22
|
+
<td><h1>Dynamic Learning Model</h1></td>
|
|
23
|
+
</tr>
|
|
24
|
+
</table>
|
|
25
|
+
|
|
26
|
+
[](https://pypi.org/project/dynamic-learning-model/)
|
|
27
|
+
[](https://pypi.org/project/dynamic-learning-model/)
|
|
28
|
+
[](LICENSE)
|
|
29
|
+
|
|
30
|
+
## Overview
|
|
31
|
+
|
|
32
|
+
The Dynamic Learning Model (DLM) is a hybrid, domain-specific AI system designed to learn, adapt, and respond intelligently to user queries. It combines natural language understanding with structured reasoning, continually improving as it is trained.
|
|
33
|
+
|
|
34
|
+
**Key capabilities include:**
|
|
35
|
+
|
|
36
|
+
- **FAQ Handling** - Learns and responds to frequently asked questions based on the knowledge it has been trained on.
|
|
37
|
+
- **Chain-of-Thought (CoT) Reasoning** - Performs clear, step-by-step logic to solve non-ambiguous arithmetic, geometric, and unit conversion problems.
|
|
38
|
+
- **Custom Knowledge Integration** - DLM is fully extensible. You can initialize it with an empty SQL database and train it with your domain-specific knowledge.
|
|
39
|
+
|
|
40
|
+
Whether you're building a student support bot, a domain-specific assistant, or a computation system, DLM offers a flexible foundation to power your intelligent applications.
|
|
41
|
+
|
|
42
|
+
## Table of Contents
|
|
43
|
+
|
|
44
|
+
- [Installation](#installation)
|
|
45
|
+
- [Required Parameters](#required-parameters)
|
|
46
|
+
- [Usage](#usage)
|
|
47
|
+
- [Important Notices](#important-notices)
|
|
48
|
+
- [License](#license)
|
|
49
|
+
- [Disclaimer](#disclaimer)
|
|
50
|
+
|
|
51
|
+
## Installation
|
|
52
|
+
|
|
53
|
+
```bash
|
|
54
|
+
pip install dynamic-learning-model
|
|
55
|
+
```
|
|
56
|
+
|
|
57
|
+
> **Requirements:** Python 3.12.0 or higher is required to use this bot in your program. All required dependencies are installed automatically with the package.
|
|
58
|
+
|
|
59
|
+
## Required Parameters
|
|
60
|
+
|
|
61
|
+
The constructor requires passing in two parameters:
|
|
62
|
+
|
|
63
|
+
1. **Bot Mode**
|
|
64
|
+
- `"learn"` - Enables training using the memory model. The bot can be updated with new information.
|
|
65
|
+
- `"apply"` - The bot automatically switches between its "compute" and "memory" models depending on the query asked.
|
|
66
|
+
2. **Database Path** - An empty (or existing) SQLite database file used for training and as the bot's memory model.
|
|
67
|
+
|
|
68
|
+
The `ask()` method also requires two parameters:
|
|
69
|
+
|
|
70
|
+
1. **Query** - e.g., `"What is the definition of FAFSA?"`
|
|
71
|
+
2. **Display Thought** - `True` to display the bot's Chain-of-Thought reasoning, or `False` to suppress it.
|
|
72
|
+
|
|
73
|
+
## Usage
|
|
74
|
+
|
|
75
|
+
**`"learn"` mode** (training queries):
|
|
76
|
+
|
|
77
|
+
```python
|
|
78
|
+
from dlm import DLM
|
|
79
|
+
|
|
80
|
+
training_bot = DLM("learn", "college_knowledge.db")
|
|
81
|
+
training_bot.ask("What is FAFSA in college?", True)
|
|
82
|
+
```
|
|
83
|
+
|
|
84
|
+
**`"apply"` mode** (deployment / production use after training):
|
|
85
|
+
|
|
86
|
+
```python
|
|
87
|
+
from dlm import DLM
|
|
88
|
+
|
|
89
|
+
commercial_bot = DLM("apply", "college_knowledge.db")
|
|
90
|
+
commercial_bot.ask("What is the difference between FAFSA and CADAA in California?", False)
|
|
91
|
+
|
|
92
|
+
# or
|
|
93
|
+
|
|
94
|
+
commercial_bot.ask("Tell me the result for the following: 5 * 5 * 5 + 5 / 5", True)
|
|
95
|
+
```
|
|
96
|
+
|
|
97
|
+
### High-Level Pipeline
|
|
98
|
+
|
|
99
|
+
|
|
100
|
+
|
|
101
|
+
## Important Notices
|
|
102
|
+
|
|
103
|
+
1. **Training data quality matters.** DLM's accuracy in `"learn"` mode depends entirely on the consistency and clarity of the question/answer pairs it's trained with. Inconsistent category labeling or vague phrasing during training can produce inaccurate or corrupted responses later.
|
|
104
|
+
2. **Database files are local and untracked.** DLM stores all trained knowledge in the SQLite file you provide. Back up this file regularly - there is no built-in cloud sync, versioning, or recovery mechanism.
|
|
105
|
+
3. **Model loading behavior.** Underlying NLP and transformer models (spaCy, HuggingFace) are lazy-loaded and shared across instances. The first call in a session may take longer due to model loading; subsequent calls are significantly faster.
|
|
106
|
+
4. **Compute mode limitations.** Chain-of-Thought computation is designed for clear, non-ambiguous arithmetic, geometric, and unit-conversion problems. Ambiguous or multi-interpretation queries may produce incorrect results - always verify outputs for critical use cases.
|
|
107
|
+
|
|
108
|
+
## License
|
|
109
|
+
|
|
110
|
+
This project is licensed under the **MIT License** - see the [LICENSE](LICENSE) file for details.
|
|
111
|
+
|
|
112
|
+
## Disclaimer
|
|
113
|
+
|
|
114
|
+
Dynamic Learning Model (DLM) is provided **"as-is"**, without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement. In no event shall the author be liable for any claim, damages, or other liability arising from the use of this software.
|
|
115
|
+
|
|
116
|
+
DLM **may produce inaccurate, incomplete, or unexpected responses**, particularly for ambiguous queries or insufficiently trained knowledge bases. **Do not rely on DLM's output for decisions involving safety, legal, medical, or financial consequences without independent verification.**
|
|
117
|
+
|
|
118
|
+
All data provided to DLM (training queries, database contents) is processed and stored **locally** on the host machine. DLM does not transmit user data externally, except for any underlying third-party model downloads (e.g. HuggingFace, spaCy) required on first run, which are subject to those providers' own terms.
|
|
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
|
|
|
4
4
|
|
|
5
5
|
[project]
|
|
6
6
|
name = "dynamic-learning-model"
|
|
7
|
-
version = "4.0.
|
|
7
|
+
version = "4.0.3"
|
|
8
8
|
description = "A Dynamic-Learning Model (DLM) chatbot with memory and compute reasoning modes."
|
|
9
9
|
readme = "README.md"
|
|
10
10
|
requires-python = ">=3.12"
|
|
@@ -1,87 +0,0 @@
|
|
|
1
|
-
Metadata-Version: 2.4
|
|
2
|
-
Name: dynamic-learning-model
|
|
3
|
-
Version: 4.0.1
|
|
4
|
-
Summary: A Dynamic-Learning Model (DLM) chatbot with memory and compute reasoning modes.
|
|
5
|
-
Home-page: https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
6
|
-
Author: Vignesh Thondikulam
|
|
7
|
-
Author-email: Vignesh Thondikulam <vignesh.tho2006@gmail.com>
|
|
8
|
-
License: MIT
|
|
9
|
-
Project-URL: Homepage, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
10
|
-
Project-URL: Repository, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
11
|
-
Requires-Python: >=3.12,<3.13
|
|
12
|
-
Description-Content-Type: text/markdown
|
|
13
|
-
License-File: LICENSE
|
|
14
|
-
Requires-Dist: spacy>=3.7
|
|
15
|
-
Requires-Dist: transformers>=4.30
|
|
16
|
-
Requires-Dist: torch>=2.0
|
|
17
|
-
Requires-Dist: better-profanity>=0.7
|
|
18
|
-
Requires-Dist: nltk>=3.8
|
|
19
|
-
Dynamic: author
|
|
20
|
-
Dynamic: home-page
|
|
21
|
-
Dynamic: license-file
|
|
22
|
-
Dynamic: requires-python
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
# Dynamic Learning Model
|
|
31
|
-
**ABOUT**:
|
|
32
|
-
|
|
33
|
-
The Dynamic Learning Model (DLM) is a hybrid AI system designed to learn, adapt, and intelligently respond to user queries. It combines natural language understanding with structured reasoning, continually improving as it is trained.
|
|
34
|
-
|
|
35
|
-
Key capabilities include:
|
|
36
|
-
|
|
37
|
-
* FAQ Handling: Learns and responds to frequently asked questions based on the knowledge it has been trained on.
|
|
38
|
-
|
|
39
|
-
* Chain-of-Thought (CoT) Reasoning: Performs clear, step-by-step logic to solve non-ambiguous arithmetic, geometric, and unit conversion problems.
|
|
40
|
-
|
|
41
|
-
* Custom Knowledge Integration: DLM is fully extensible. You can initialize it with an empty SQL database and train it with your domain-specific knowledge.
|
|
42
|
-
|
|
43
|
-
Whether you're building a student support bot, a domain-specific assistant, or a computation system, DLM offers a flexible foundation to power your intelligent applications
|
|
44
|
-
|
|
45
|
-
**REQUIRED PARAMETERS**:
|
|
46
|
-
* The constructor requires passing in two parameters:
|
|
47
|
-
- Bot Mode:
|
|
48
|
-
- 'learn' = Enables training using the memory model. The bot can be updated with new information
|
|
49
|
-
- 'apply' = The bot automatically switches between its "compute" and "memory" model depending on the query asked
|
|
50
|
-
- Empty SQL Database for training the bot with queries and for the memory model
|
|
51
|
-
* The ask() method also requires passing in two parameters:
|
|
52
|
-
- Query: "What is the definition of FAFSA" (as an example)
|
|
53
|
-
- Display Thought: "True" to allow the bot's Chain of Thought to be displayed, or else "False"
|
|
54
|
-
|
|
55
|
-
**GET STARTED**:
|
|
56
|
-
* To install, run:
|
|
57
|
-
```bash
|
|
58
|
-
pip install dynamic-learning-model
|
|
59
|
-
```
|
|
60
|
-
* ***Python 3.12.0 is required to use this bot in your program.***
|
|
61
|
-
|
|
62
|
-
('learn' mode [training queries])
|
|
63
|
-
```python
|
|
64
|
-
from dlm import DLM
|
|
65
|
-
|
|
66
|
-
training_bot = DLM("learn", "college_knowledge.db")
|
|
67
|
-
|
|
68
|
-
training_bot.ask("What is FAFSA in college?", True)
|
|
69
|
-
```
|
|
70
|
-
|
|
71
|
-
('apply' mode [deployment/production use after training])
|
|
72
|
-
```python
|
|
73
|
-
from dlm import DLM
|
|
74
|
-
|
|
75
|
-
commercial_bot = DLM("apply", "college_knowledge.db")
|
|
76
|
-
|
|
77
|
-
commercial_bot.ask("What is the difference between FAFSA and CADAA in California?", False)
|
|
78
|
-
|
|
79
|
-
# or
|
|
80
|
-
|
|
81
|
-
commercial_bot.ask("Tell me the result for the following: 5 * 5 * 5 + 5 / 5", True)
|
|
82
|
-
```
|
|
83
|
-
|
|
84
|
-
**HIGH-LEVEL PIPELINE VISUAL**:
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
|
|
@@ -1,64 +0,0 @@
|
|
|
1
|
-
|
|
2
|
-
|
|
3
|
-
|
|
4
|
-
|
|
5
|
-
|
|
6
|
-
|
|
7
|
-
# Dynamic Learning Model
|
|
8
|
-
**ABOUT**:
|
|
9
|
-
|
|
10
|
-
The Dynamic Learning Model (DLM) is a hybrid AI system designed to learn, adapt, and intelligently respond to user queries. It combines natural language understanding with structured reasoning, continually improving as it is trained.
|
|
11
|
-
|
|
12
|
-
Key capabilities include:
|
|
13
|
-
|
|
14
|
-
* FAQ Handling: Learns and responds to frequently asked questions based on the knowledge it has been trained on.
|
|
15
|
-
|
|
16
|
-
* Chain-of-Thought (CoT) Reasoning: Performs clear, step-by-step logic to solve non-ambiguous arithmetic, geometric, and unit conversion problems.
|
|
17
|
-
|
|
18
|
-
* Custom Knowledge Integration: DLM is fully extensible. You can initialize it with an empty SQL database and train it with your domain-specific knowledge.
|
|
19
|
-
|
|
20
|
-
Whether you're building a student support bot, a domain-specific assistant, or a computation system, DLM offers a flexible foundation to power your intelligent applications
|
|
21
|
-
|
|
22
|
-
**REQUIRED PARAMETERS**:
|
|
23
|
-
* The constructor requires passing in two parameters:
|
|
24
|
-
- Bot Mode:
|
|
25
|
-
- 'learn' = Enables training using the memory model. The bot can be updated with new information
|
|
26
|
-
- 'apply' = The bot automatically switches between its "compute" and "memory" model depending on the query asked
|
|
27
|
-
- Empty SQL Database for training the bot with queries and for the memory model
|
|
28
|
-
* The ask() method also requires passing in two parameters:
|
|
29
|
-
- Query: "What is the definition of FAFSA" (as an example)
|
|
30
|
-
- Display Thought: "True" to allow the bot's Chain of Thought to be displayed, or else "False"
|
|
31
|
-
|
|
32
|
-
**GET STARTED**:
|
|
33
|
-
* To install, run:
|
|
34
|
-
```bash
|
|
35
|
-
pip install dynamic-learning-model
|
|
36
|
-
```
|
|
37
|
-
* ***Python 3.12.0 is required to use this bot in your program.***
|
|
38
|
-
|
|
39
|
-
('learn' mode [training queries])
|
|
40
|
-
```python
|
|
41
|
-
from dlm import DLM
|
|
42
|
-
|
|
43
|
-
training_bot = DLM("learn", "college_knowledge.db")
|
|
44
|
-
|
|
45
|
-
training_bot.ask("What is FAFSA in college?", True)
|
|
46
|
-
```
|
|
47
|
-
|
|
48
|
-
('apply' mode [deployment/production use after training])
|
|
49
|
-
```python
|
|
50
|
-
from dlm import DLM
|
|
51
|
-
|
|
52
|
-
commercial_bot = DLM("apply", "college_knowledge.db")
|
|
53
|
-
|
|
54
|
-
commercial_bot.ask("What is the difference between FAFSA and CADAA in California?", False)
|
|
55
|
-
|
|
56
|
-
# or
|
|
57
|
-
|
|
58
|
-
commercial_bot.ask("Tell me the result for the following: 5 * 5 * 5 + 5 / 5", True)
|
|
59
|
-
```
|
|
60
|
-
|
|
61
|
-
**HIGH-LEVEL PIPELINE VISUAL**:
|
|
62
|
-
|
|
63
|
-
|
|
64
|
-
|
|
@@ -1,87 +0,0 @@
|
|
|
1
|
-
Metadata-Version: 2.4
|
|
2
|
-
Name: dynamic-learning-model
|
|
3
|
-
Version: 4.0.1
|
|
4
|
-
Summary: A Dynamic-Learning Model (DLM) chatbot with memory and compute reasoning modes.
|
|
5
|
-
Home-page: https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
6
|
-
Author: Vignesh Thondikulam
|
|
7
|
-
Author-email: Vignesh Thondikulam <vignesh.tho2006@gmail.com>
|
|
8
|
-
License: MIT
|
|
9
|
-
Project-URL: Homepage, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
10
|
-
Project-URL: Repository, https://github.com/VigneshT24/Dynamic_Learning_Model
|
|
11
|
-
Requires-Python: >=3.12,<3.13
|
|
12
|
-
Description-Content-Type: text/markdown
|
|
13
|
-
License-File: LICENSE
|
|
14
|
-
Requires-Dist: spacy>=3.7
|
|
15
|
-
Requires-Dist: transformers>=4.30
|
|
16
|
-
Requires-Dist: torch>=2.0
|
|
17
|
-
Requires-Dist: better-profanity>=0.7
|
|
18
|
-
Requires-Dist: nltk>=3.8
|
|
19
|
-
Dynamic: author
|
|
20
|
-
Dynamic: home-page
|
|
21
|
-
Dynamic: license-file
|
|
22
|
-
Dynamic: requires-python
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
# Dynamic Learning Model
|
|
31
|
-
**ABOUT**:
|
|
32
|
-
|
|
33
|
-
The Dynamic Learning Model (DLM) is a hybrid AI system designed to learn, adapt, and intelligently respond to user queries. It combines natural language understanding with structured reasoning, continually improving as it is trained.
|
|
34
|
-
|
|
35
|
-
Key capabilities include:
|
|
36
|
-
|
|
37
|
-
* FAQ Handling: Learns and responds to frequently asked questions based on the knowledge it has been trained on.
|
|
38
|
-
|
|
39
|
-
* Chain-of-Thought (CoT) Reasoning: Performs clear, step-by-step logic to solve non-ambiguous arithmetic, geometric, and unit conversion problems.
|
|
40
|
-
|
|
41
|
-
* Custom Knowledge Integration: DLM is fully extensible. You can initialize it with an empty SQL database and train it with your domain-specific knowledge.
|
|
42
|
-
|
|
43
|
-
Whether you're building a student support bot, a domain-specific assistant, or a computation system, DLM offers a flexible foundation to power your intelligent applications
|
|
44
|
-
|
|
45
|
-
**REQUIRED PARAMETERS**:
|
|
46
|
-
* The constructor requires passing in two parameters:
|
|
47
|
-
- Bot Mode:
|
|
48
|
-
- 'learn' = Enables training using the memory model. The bot can be updated with new information
|
|
49
|
-
- 'apply' = The bot automatically switches between its "compute" and "memory" model depending on the query asked
|
|
50
|
-
- Empty SQL Database for training the bot with queries and for the memory model
|
|
51
|
-
* The ask() method also requires passing in two parameters:
|
|
52
|
-
- Query: "What is the definition of FAFSA" (as an example)
|
|
53
|
-
- Display Thought: "True" to allow the bot's Chain of Thought to be displayed, or else "False"
|
|
54
|
-
|
|
55
|
-
**GET STARTED**:
|
|
56
|
-
* To install, run:
|
|
57
|
-
```bash
|
|
58
|
-
pip install dynamic-learning-model
|
|
59
|
-
```
|
|
60
|
-
* ***Python 3.12.0 is required to use this bot in your program.***
|
|
61
|
-
|
|
62
|
-
('learn' mode [training queries])
|
|
63
|
-
```python
|
|
64
|
-
from dlm import DLM
|
|
65
|
-
|
|
66
|
-
training_bot = DLM("learn", "college_knowledge.db")
|
|
67
|
-
|
|
68
|
-
training_bot.ask("What is FAFSA in college?", True)
|
|
69
|
-
```
|
|
70
|
-
|
|
71
|
-
('apply' mode [deployment/production use after training])
|
|
72
|
-
```python
|
|
73
|
-
from dlm import DLM
|
|
74
|
-
|
|
75
|
-
commercial_bot = DLM("apply", "college_knowledge.db")
|
|
76
|
-
|
|
77
|
-
commercial_bot.ask("What is the difference between FAFSA and CADAA in California?", False)
|
|
78
|
-
|
|
79
|
-
# or
|
|
80
|
-
|
|
81
|
-
commercial_bot.ask("Tell me the result for the following: 5 * 5 * 5 + 5 / 5", True)
|
|
82
|
-
```
|
|
83
|
-
|
|
84
|
-
**HIGH-LEVEL PIPELINE VISUAL**:
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
|
|
@@ -1,32 +0,0 @@
|
|
|
1
|
-
from setuptools import setup, find_packages
|
|
2
|
-
|
|
3
|
-
with open('README.md', 'r', encoding='utf-8') as f:
|
|
4
|
-
long_description = f.read()
|
|
5
|
-
|
|
6
|
-
setup(
|
|
7
|
-
name='dynamic-learning-model',
|
|
8
|
-
version='3.5.2',
|
|
9
|
-
author='Vignesh Thondikulam',
|
|
10
|
-
author_email='vignesh.tho2006@gmail.com',
|
|
11
|
-
description='A Dynamic Learning Model for processing NLP queries using hybrid AI and reasoning.',
|
|
12
|
-
long_description=long_description,
|
|
13
|
-
long_description_content_type='text/markdown',
|
|
14
|
-
url='https://github.com/VigneshT24/Dynamic_Learning_Model',
|
|
15
|
-
packages=find_packages(),
|
|
16
|
-
include_package_data=True,
|
|
17
|
-
install_requires=[
|
|
18
|
-
'nltk',
|
|
19
|
-
'spacy',
|
|
20
|
-
'better_profanity',
|
|
21
|
-
'word2number',
|
|
22
|
-
'transformers',
|
|
23
|
-
'hf_xet'
|
|
24
|
-
],
|
|
25
|
-
classifiers=[
|
|
26
|
-
'Programming Language :: Python :: 3',
|
|
27
|
-
'Operating System :: OS Independent',
|
|
28
|
-
'License :: OSI Approved :: MIT License',
|
|
29
|
-
],
|
|
30
|
-
python_requires='>=3.12,<3.13',
|
|
31
|
-
license='MIT',
|
|
32
|
-
)
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|