Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
A neural network to generate captions for an image using CNN and RNN with BEAM Search.
| Date | Stars |
|---|---|
| 2026-07-31 | 306 |
| 2026-08-05 | 306 |
| 2026-08-06 | 306 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
## Image Caption Generator [](https://github.com/dabasajay/Image-Caption-Generator/issues) [](https://github.com/dabasajay/Image-Caption-Generator/network) [](https://github.com/dabasajay/Image-Caption-Generator/stargazers) [](https://dabasajay.github.io/) A neural network to generate captions for an image using CNN and RNN with BEAM Search. <p align="center"> <strong>Examples</strong> </p> <p align="center"> <img src="https://cdn-images-1.medium.com/max/1600/1*6BFOIdSHlk24Z3DFEakvnQ.png" width="85%" title="Example of Image Captioning" alt="Example of Image Captioning"> </p> <p align="center"> Image Credits : <a href="https://towardsdatascience.com/image-captioning-in-deep-learning-9cd23fb4d8d2">Towardsdatascience</a> </p> ## Table of Contents 1. [Requirements](#1-requirements) 2. [Training parameters and results](#2-training-parameters-and-results) 3. [Generated Captions on Test Images](#3-generated-captions-on-test-images) 4. [Procedure to Train Model](#4-procedure-to-train-model) 5. [Procedure to Test on new images](#5-procedure-to-test-on-new-images) 6. [Configurations (config.py)](#6-configurations-configpy) 7. [Frequently encountered problems](#7-frequently-encountered-problems) 8. [TODO](#8-todo) 9. [References](#9-references) ## 1. Requirements Recommended System Requirements to train model. <ul type="square"> <li>A good CPU and a GPU with atleast 8GB memory</li> <li>Atleast 8GB of RAM</li> <li>Active internet connection so that keras can download inceptionv3/vgg16 model weights</li> </ul> Required libraries for Python along with their version numbers used while making & testing of this project <ul type="square"> <li>Python - 3.6.7</li> <li>Numpy - 1.16.4</li> <li>Tensorflow - 1.13.1</li> <li>Keras - 2.2.4</li> <li>nltk - 3.2.5</li> <li>PIL - 4.3.0</li> <li>Matplotlib - 3.0.3</li> <li>tqdm - 4.28.1</li> </ul> <strong>Flickr8k Dataset:</strong> <a href="https://forms.illinois.edu/sec/1713398">Dataset Request Form</a> <strong>UPDATE (April/2019):</strong> The official site seems to have been taken down (although the form still works). Here are some direct download links: <ul type="square"> <li><a href="https://github.com/jbrownlee/Datasets/releases/download/Flickr8k/Flickr8k_Dataset.zip">Flickr8k_Dataset</a></li> <li><a href="https://github.com/jbrownlee/Datasets/releases/download/Flickr8k/Flickr8k_text.zip">Flickr8k_text</a></li> Download Link Credits:<a href="https://machinelearningmastery.com/develop-a-deep-learning-caption-generation-model-in-python/"> Jason Brownlee</a> </ul> <strong>Important:</strong> After downloading the dataset, put the reqired files in train_val_data folder ## 2. Training parameters and results #### NOTE - `batch_size=64` took ~14GB GPU memory in case of *InceptionV3 + AlternativeRNN* and *VGG16 + AlternativeRNN* - `batch_size=64` took ~8GB GPU memory in case of *InceptionV3 + RNN* and *VGG16 + RNN* - **If you're low on memory**, use google colab or reduce batch size - In case of BEAM Search, `loss` and `val_loss` are same as in case of argmax since the model is same | Model & Config | Argmax | BEAM Search | | :--- | :--- | :--- | | **InceptionV3 + AlternativeRNN** <ul><li>Epochs = 20</li><li>Batch Size = 64</li><li>Optimizer = Adam</li></ul> |<ul>**Crossentropy loss**<br>*(Lower the better)*<li>loss(train_loss): 2.4050</li><li>val_loss: 3.0527</li>**BLEU Scores on Validation data**<br>*(Higher the better)*<li>BLEU-1: 0.596818</li><li>BLEU-2: 0.356009</li><li>BLEU-3: 0.252489</li><li>BLEU-4: 0.129536</li></ul> |<ul>**k = 3**<br><br>**BLEU Scores on Validation data**<br>*(Higher the better)*<li>BLEU
Excerpt of 10,268 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:a2af78dbe4b839dd, llm:Repository description and README: 'A neural network to generate captions for an image using CNN and RNN with BEAM Search.' Topics: image-captioning, caption-generation, cnn-keras, lstm, inception-v3, flickr-8k, beam-search, bleu-score.
matched fp:a2af78dbe4b839dd, llm:Repository description and README: 'A neural network to generate captions for an image using CNN and RNN with BEAM Search.' Topics: image-captioning, caption-generation, cnn-keras, lstm, inception-v3, flickr-8k, beam-search, bleu-score.
matched fp:a2af78dbe4b839dd, llm:Repository description and README: 'A neural network to generate captions for an image using CNN and RNN with BEAM Search.' Topics: image-captioning, caption-generation, cnn-keras, lstm, inception-v3, flickr-8k, beam-search, bleu-score.
matched fp:a2af78dbe4b839dd, llm:Repository description and README: 'A neural network to generate captions for an image using CNN and RNN with BEAM Search.' Topics: image-captioning, caption-generation, cnn-keras, lstm, inception-v3, flickr-8k, beam-search, bleu-score.