L1aoXingyu / Char Rnn Pytorch
使用PyTorch实现Char RNN生成古诗和周杰伦的歌词
Stars: ✭ 114
Projects that are alternatives of or similar to Char Rnn Pytorch
Attentive Neural Processes
implementing "recurrent attentive neural processes" to forecast power usage (w. LSTM baseline, MCDropout)
Stars: ✭ 33 (-71.05%)
Mutual labels: jupyter-notebook, rnn
Bitcoin Price Prediction Using Lstm
Bitcoin price Prediction ( Time Series ) using LSTM Recurrent neural network
Stars: ✭ 67 (-41.23%)
Mutual labels: jupyter-notebook, rnn
Stockpriceprediction
Stock Price Prediction using Machine Learning Techniques
Stars: ✭ 700 (+514.04%)
Mutual labels: jupyter-notebook, rnn
Pytorch Pos Tagging
A tutorial on how to implement models for part-of-speech tagging using PyTorch and TorchText.
Stars: ✭ 96 (-15.79%)
Mutual labels: jupyter-notebook, rnn
Tensorflow cookbook
Code for Tensorflow Machine Learning Cookbook
Stars: ✭ 5,984 (+5149.12%)
Mutual labels: jupyter-notebook, rnn
Tensorflow Tutorials For Time Series
TensorFlow Tutorial for Time Series Prediction
Stars: ✭ 1,067 (+835.96%)
Mutual labels: jupyter-notebook, rnn
Headlines
Automatically generate headlines to short articles
Stars: ✭ 516 (+352.63%)
Mutual labels: jupyter-notebook, rnn
Speech Emotion Recognition
Detecting emotions using MFCC features of human speech using Deep Learning
Stars: ✭ 89 (-21.93%)
Mutual labels: jupyter-notebook, rnn
Lstm chem
Implementation of the paper - Generative Recurrent Networks for De Novo Drug Design.
Stars: ✭ 87 (-23.68%)
Mutual labels: jupyter-notebook, rnn
Telemanom
A framework for using LSTMs to detect anomalies in multivariate time series data. Includes spacecraft anomaly data and experiments from the Mars Science Laboratory and SMAP missions.
Stars: ✭ 589 (+416.67%)
Mutual labels: jupyter-notebook, rnn
Codesearchnet
Datasets, tools, and benchmarks for representation learning of code.
Stars: ✭ 1,378 (+1108.77%)
Mutual labels: jupyter-notebook, rnn
Video Classification
Tutorial for video classification/ action recognition using 3D CNN/ CNN+RNN on UCF101
Stars: ✭ 543 (+376.32%)
Mutual labels: jupyter-notebook, rnn
Lstm Sentiment Analysis
Sentiment Analysis with LSTMs in Tensorflow
Stars: ✭ 886 (+677.19%)
Mutual labels: jupyter-notebook, rnn
Deeplearning
深度学习入门教程, 优秀文章, Deep Learning Tutorial
Stars: ✭ 6,783 (+5850%)
Mutual labels: jupyter-notebook, rnn
Rnn Notebooks
RNN(SimpleRNN, LSTM, GRU) Tensorflow2.0 & Keras Notebooks (Workshop materials)
Stars: ✭ 48 (-57.89%)
Mutual labels: jupyter-notebook, rnn
Easy Deep Learning With Keras
Keras tutorial for beginners (using TF backend)
Stars: ✭ 367 (+221.93%)
Mutual labels: jupyter-notebook, rnn
Tsai
Time series Timeseries Deep Learning Pytorch fastai - State-of-the-art Deep Learning with Time Series and Sequences in Pytorch / fastai
Stars: ✭ 407 (+257.02%)
Mutual labels: jupyter-notebook, rnn
Machine Learning
My Attempt(s) In The World Of ML/DL....
Stars: ✭ 78 (-31.58%)
Mutual labels: jupyter-notebook, rnn
Pytorch Learners Tutorial
PyTorch tutorial for learners
Stars: ✭ 97 (-14.91%)
Mutual labels: jupyter-notebook, rnn
Char-RNN-PyTorch
使用字符级别的RNN进行文本生成,使用PyTorch框架。Gluon实现
Requirements
按照 pytorch 官网安装 pytorch,将 mxtorch 下载下来,放到根目录,安装 tensorboardX 实现 tensorboard 可视化
\Char-RNN-PyTorch
\mxtorch
\data
\dataset
\models
config.py
main.py
训练模型
所有的配置文件都放在 config.py 里面,通过下面的代码来训练模型
python main.py train
也可以在终端修改配置
python main.py train \
--txt='./dataset/poetry.txt' \ # 训练用的txt文本
--batch=128 \ # batch_size
--max_epoch=300 \
--len=30 \ # 输入RNN的序列长度
--max_vocab=5000 \ # 最大的字符数量
--embed_dim=512 \ # 词向量的维度
--hidden_size=512 \ # 网络的输出维度
--num_layers=2 \ # RNN的层数
--dropout=0.5
如果希望使用训练好的网络进行文本生成,使用下面的代码
python main.py predict \
--begin='天青色等烟雨' \ # 生成文本的开始,可以是一个字符,也可以一段话
--predict_len=100 \ # 希望生成文本的长度
--load_model='./checkpoints/CharRNN_best_model.pth' # 读取训练模型的位置
Result
如果使用古诗的数据集进行训练,可以得到下面的结果
天青色等烟雨翩 黄望堪魄弦夜 逐奏文明际天月辉 豪天明月天趣 天外何山重满 遥天明上天 心空游无拂天外空寂室叨
如果使用周杰伦的歌词作为训练集,可以得到下面的结果
这感觉得可能 我这玻童来 城堡药比生对这些年风天 脚剧飘逐在尘里里步的路 麦缘日下一经经 听觉得远回白择
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].