Comments (5)
ralscha
commented on September 17, 2024
You are right. stats is not used. It's just for information and contains some statistics about the models. This information is not needed for running the model and thus not saved.
You could add console.log(stats) and see what it contains
from blog2019.
MinerNJ
commented on September 17, 2024
Ah ok thank you! Where is the accuracy of the model recorded at? Following your blog on Machine Learning with BrainJS shows your accuracy, but I wasn't able to find it in the model.json. Also, I apologize if this is not the place to ask that question.
from blog2019.
ralscha
commented on September 17, 2024
Good question. I don't remember from where I got the accuracy. model.json only contains the weights and when you print stats with console.log(stats); you get this information.
{
avgs: {
trainTime: 836414.5,
testTime: 16537.25,
iterations: 3,
error: 0.004108120867492793
},
stats: { total: 210000, testSize: 52500, trainSize: 157500 },
sets: [
{
error: 0.00411777707088103,
misclasses: [Array],
total: 52500,
trainTime: 735011,
testTime: 16787,
iterations: 3,
learningRate: 0.3,
hiddenLayers: [Array],
network: [Object]
},
{
error: 0.0042104415590232215,
misclasses: [Array],
total: 52500,
trainTime: 937039,
testTime: 16581,
iterations: 3,
learningRate: 0.3,
hiddenLayers: [Array],
network: [Object]
},
{
error: 0.004051989221686814,
misclasses: [Array],
total: 52500,
trainTime: 828726,
testTime: 16495,
iterations: 3,
learningRate: 0.3,
hiddenLayers: [Array],
network: [Object]
},
{
error: 0.004052275618380105,
misclasses: [Array],
total: 52500,
trainTime: 844882,
testTime: 16286,
iterations: 3,
learningRate: 0.3,
hiddenLayers: [Array],
network: [Object]
}
]
}
from blog2019.
ralscha
commented on September 17, 2024
Oh. I remember. I wrote an additional program that tested the model:
https://github.com/ralscha/blog2019/blob/master/mnistjs/train/run.js
Run: node run.js
Output:
Total : 10000
Number of errors: 180
Accuracy: : 98.2 %
from blog2019.
MinerNJ
commented on September 17, 2024
Ah of course. I'm sorry, I should have gone through the source code more
closely. Thank you for all your help!
…On Tue, May 21, 2019, 2:33 PM Ralph Schaer ***@***.***> wrote:
Oh. I remember. I wrote an additional program that tested the model:
https://github.com/ralscha/blog2019/blob/master/mnistjs/train/run.js
Run: node run.js
Output:
Total : 10000
Number of errors: 180
Accuracy: : 98.2 %
—
You are receiving this because you modified the open/close state.
Reply to this email directly, view it on GitHub
<#3?email_source=notifications&email_token=ALEEYHB6QKW4S35P5CCI76DPWQ573A5CNFSM4HOLDUH2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4ZIPI#issuecomment-494507069>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ALEEYHFC2GTDPOXIY2V4I23PWQ573ANCNFSM4HOLDUHQ>
.
from blog2019.
Related Issues (7)
- Could not autowire. There is more than one bean of 'HazelcastInstance' type. HOT 3
- non-compliant form encoding HOT 2
- run these model locally HOT 1
- This implementation is __INSECURE__. HOT 9
- CAP camera api will open camera even though try PICTURE
- not able to make push notifications pop up on screen of android phone HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from blog2019.