kandi X-RAY | DocDex Summary
kandi X-RAY | DocDex Summary
Warning! This software is in beta, it is extremely likely there'll be bugs, and some ugly things. Please report bugs, and make suggestions via the issues tab. Documentation Index (DocDex) is a utility API which scans javadocs on the web, caches everything into a mixture of memory, file, and database, then provides a JSON API to easily (and quickly!) access the information.
Top functions reviewed by kandi - BETA
- Provides a map of objects indexed by the given javadoc .
- Get methods .
- Returns a string representation of an object .
- Escapes the string so that it can be used as an SQL source .
- Processes a bot command .
- Loads from raw server .
- Determine the distance between two byte arrays .
- Processes a paginated message .
- Generic action handler .
- Downloads a set of URIs .
DocDex Key Features
DocDex Examples and Code Snippets
Trending Discussions on DocDex
I have written tensorflow code based on:
but using precomputed word embeddings from the GoogleNews word2vec 300 dimension model.
I created my own data from the UCML News Aggregator Dataset in which I parsed the content of the news articles and have created my own labels.
Due to the size of the articles I use TF-IDF to filter out the top 120 words per article and embed those into 300 dimensions.
When I run the CNN I created regardless of the hyper parameters it converges to a small general accuracy, around 38%.
Hyper parameters changed:
Various filter sizes:
I've tried a single filter of 1,2,3 Combinations of filters [3,4,5], [1,3,4]
I've varied this from very low to very high, very low doesn't converge to 38% but anything between 0.0001 and 0.4 does.
Tried many ranges between 5 and 100.
Weight and Bias Initialization:
Set stddev of weights between 0.4 and 0.01. Set bias initial values between 0 and 0.1. Tried using the xavier initializer for the conv2d weights.
I have only tried on two partial data sets, one with 15 000 training data, and the other on the 5000 test data. In total I have 263 000 data to train on. There is no accuracy difference whether trained and evaluated on the 15 000 training data or by using the 5000 test data as the training data (to save testing time).
I've run successful classifications on the 15 000 / 5000 split using a feed forward network with a BoW input (93% accurate), TF-IDF with SVM (92%), and TF-IDF with Native Bayes (91.5%). So I don't think it is the data.
What does this imply? Is the model just a poor model for this task? Is there an error in my work?
I feel like my do_eval function is incorrect to evaluate the accuracy / loss over an epoch of the data:...
ANSWERAnswered 2017-Sep-20 at 00:19
Turns out my error was in the creation of the input matrix.
No vulnerabilities reported
You can use DocDex like any standard Java library. Please include the the jar files in your classpath. You can also use any IDE and you can run and debug the DocDex component as you would do with any other Java program. Best practice is to use a build tool that supports dependency management such as Maven or Gradle. For Maven installation, please refer maven.apache.org. For Gradle installation, please refer gradle.org .
Reuse Trending Solutions
Subscribe to our newsletter for trending solutions and developer bootcamps
Share this Page