Skip to content

Commit ebc0b6f

Browse files
hemangjoshi37ahemangjoshi37a
committed
new
1 parent 974cba4 commit ebc0b6f

23 files changed

+2984
-3533
lines changed

‎Inference/inference_notebook.ipynb

+187-41
Large diffs are not rendered by default.

‎Inference/notebook62326ade97 (1).ipynb

+150-107
Large diffs are not rendered by default.

‎Training/TrendMaster_1X.ipynb

+666
Large diffs are not rendered by default.

‎Training/TrendMaster_5X.ipynb

+1,292
Large diffs are not rendered by default.

‎Training/best_model_multi10.pt

-355 KB
Binary file not shown.

‎Training/best_model_multi11.pt

-354 KB
Binary file not shown.

‎docs/api_reference.md

+164
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,164 @@
1+
# TrendMaster API Reference
2+
3+
This document provides detailed information about the classes and functions in the TrendMaster package.
4+
5+
## Table of Contents
6+
7+
1. [DataLoader](#dataloader)
8+
2. [TransAm (Model)](#transam-model)
9+
3. [Trainer](#trainer)
10+
4. [Inferencer](#inferencer)
11+
5. [Utility Functions](#utility-functions)
12+
13+
## DataLoader
14+
15+
The `DataLoader` class is responsible for loading and preprocessing stock data.
16+
17+
### Methods
18+
19+
#### `__init__(self)`
20+
Initializes the DataLoader object.
21+
22+
#### `authenticate(self, user_id=None, password=None, twofa=None)`
23+
Authenticates with the Zerodha API.
24+
25+
- Parameters:
26+
- `user_id` (str, optional): Zerodha user ID
27+
- `password` (str, optional): Zerodha password
28+
- `twofa` (str, optional): Zerodha two-factor authentication code
29+
- Returns: Authenticated Zerodha kite instance
30+
31+
#### `get_stock_data(self, symbol, from_date, to_date, interval='minute')`
32+
Fetches stock data for a given symbol.
33+
34+
- Parameters:
35+
- `symbol` (str): Stock symbol
36+
- `from_date` (str): Start date in 'YYYY-MM-DD' format
37+
- `to_date` (str): End date in 'YYYY-MM-DD' format
38+
- `interval` (str, optional): Data interval (default: 'minute')
39+
- Returns: pandas DataFrame containing stock data
40+
41+
#### `preprocess_data(self, data, column='close')`
42+
Preprocesses the data for model input.
43+
44+
- Parameters:
45+
- `data` (pandas.DataFrame): Stock data
46+
- `column` (str, optional): Column to preprocess (default: 'close')
47+
- Returns: Preprocessed numpy array
48+
49+
#### `create_sequences(self, data, input_window, output_window)`
50+
Creates input-output sequences for training.
51+
52+
- Parameters:
53+
- `data` (numpy.array): Preprocessed data
54+
- `input_window` (int): Number of input time steps
55+
- `output_window` (int): Number of output time steps
56+
- Returns: List of (input_sequence, target_sequence) tuples
57+
58+
## TransAm (Model)
59+
60+
The `TransAm` class implements the Transformer model for stock price prediction.
61+
62+
### Methods
63+
64+
#### `__init__(self, feature_size=30, num_layers=2, dropout=0.2)`
65+
Initializes the TransAm model.
66+
67+
- Parameters:
68+
- `feature_size` (int, optional): Size of input features (default: 30)
69+
- `num_layers` (int, optional): Number of transformer layers (default: 2)
70+
- `dropout` (float, optional): Dropout rate (default: 0.2)
71+
72+
#### `forward(self, src)`
73+
Performs a forward pass through the model.
74+
75+
- Parameters:
76+
- `src` (torch.Tensor): Input tensor
77+
- Returns: Output tensor
78+
79+
## Trainer
80+
81+
The `Trainer` class handles model training and validation.
82+
83+
### Methods
84+
85+
#### `__init__(self, model, device, learning_rate=0.001)`
86+
Initializes the Trainer object.
87+
88+
- Parameters:
89+
- `model` (TransAm): The model to train
90+
- `device` (torch.device): Device to use for training
91+
- `learning_rate` (float, optional): Learning rate (default: 0.001)
92+
93+
#### `train(self, train_data, val_data, epochs, batch_size)`
94+
Trains the model.
95+
96+
- Parameters:
97+
- `train_data` (list): Training data sequences
98+
- `val_data` (list): Validation data sequences
99+
- `epochs` (int): Number of training epochs
100+
- `batch_size` (int): Batch size for training
101+
- Returns: Lists of training and validation losses
102+
103+
#### `validate(self, val_data, batch_size)`
104+
Validates the model on the provided data.
105+
106+
- Parameters:
107+
- `val_data` (list): Validation data sequences
108+
- `batch_size` (int): Batch size for validation
109+
- Returns: Validation loss
110+
111+
## Inferencer
112+
113+
The `Inferencer` class handles model inference and evaluation.
114+
115+
### Methods
116+
117+
#### `__init__(self, model, device, data_loader)`
118+
Initializes the Inferencer object.
119+
120+
- Parameters:
121+
- `model` (TransAm): Trained model
122+
- `device` (torch.device): Device to use for inference
123+
- `data_loader` (DataLoader): DataLoader instance
124+
125+
#### `predict(self, symbol, from_date, to_date, input_window, future_steps)`
126+
Makes predictions for future stock prices.
127+
128+
- Parameters:
129+
- `symbol` (str): Stock symbol
130+
- `from_date` (str): Start date for historical data
131+
- `to_date` (str): End date for historical data
132+
- `input_window` (int): Number of input time steps
133+
- `future_steps` (int): Number of future time steps to predict
134+
- Returns: DataFrame with predicted prices
135+
136+
#### `evaluate(self, test_data, batch_size)`
137+
Evaluates the model on test data.
138+
139+
- Parameters:
140+
- `test_data` (list): Test data sequences
141+
- `batch_size` (int): Batch size for evaluation
142+
- Returns: Test loss
143+
144+
## Utility Functions
145+
146+
### `set_seed(seed)`
147+
Sets random seed for reproducibility.
148+
149+
- Parameters:
150+
- `seed` (int): Random seed
151+
152+
### `plot_results(train_losses, val_losses)`
153+
Plots training and validation losses.
154+
155+
- Parameters:
156+
- `train_losses` (list): Training losses
157+
- `val_losses` (list): Validation losses
158+
159+
### `plot_predictions(actual, predictions)`
160+
Plots actual vs predicted stock prices.
161+
162+
- Parameters:
163+
- `actual` (pandas.Series): Actual stock prices
164+
- `predictions` (pandas.DataFrame): Predicted stock prices

‎docs/index.md

+21
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,21 @@
1+
# TrendMaster Documentation
2+
3+
Welcome to the TrendMaster documentation. TrendMaster is a powerful stock price prediction tool that uses Transformer deep learning architecture to provide accurate forecasts.
4+
5+
## Table of Contents
6+
7+
1. [Installation](installation.md)
8+
2. [Quick Start](quickstart.md)
9+
3. [API Reference](api_reference.md)
10+
11+
## Overview
12+
13+
TrendMaster leverages the power of Transformer models to analyze historical stock data and make predictions about future price movements. It provides an easy-to-use interface for data loading, model training, and inference.
14+
15+
Key features:
16+
- Integration with Zerodha for real-time stock data
17+
- Customizable Transformer model architecture
18+
- Flexible training and inference pipelines
19+
- Visualization tools for analysis and results
20+
21+
Get started by following our [Installation Guide](installation.md) and then check out the [Quick Start](quickstart.md) tutorial to begin making predictions right away!

‎docs/installation.md

+43
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,43 @@
1+
# Installation
2+
3+
Follow these steps to install TrendMaster and its dependencies.
4+
5+
## Prerequisites
6+
7+
- Python 3.7 or higher
8+
- pip (Python package installer)
9+
10+
## Installing TrendMaster
11+
12+
You can install TrendMaster using pip:
13+
14+
```bash
15+
pip install trendmaster
16+
```
17+
18+
This will install TrendMaster and all its required dependencies.
19+
20+
## Installing from source
21+
22+
If you want to install the latest development version, you can install directly from the GitHub repository:
23+
24+
```bash
25+
git clone https://github.com/hemangjoshi37a/TrendMaster.git
26+
cd TrendMaster
27+
pip install -e .
28+
```
29+
30+
## Verifying the installation
31+
32+
After installation, you can verify that TrendMaster is installed correctly by running:
33+
34+
```python
35+
import trendmaster
36+
print(trendmaster.__version__)
37+
```
38+
39+
This should print the version number of TrendMaster.
40+
41+
## Next steps
42+
43+
Now that you have TrendMaster installed, check out the [Quick Start](quickstart.md) guide to begin using the library!

‎docs/quickstart.md

+7
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
1+
# Quick Start Guide
2+
3+
This guide will help you get started with TrendMaster quickly. We'll cover the basics of data loading, model training, and making predictions.
4+
5+
## Basic Usage
6+
7+
Follow the code in the examples folder

0 commit comments

Comments
 (0)