ivonatau / emotionaldan Goto Github PK
View Code? Open in Web Editor NEWDeep Neural Network for joint emotion classification and landmark localization.
Home Page: https://arxiv.org/abs/1810.10529
Deep Neural Network for joint emotion classification and landmark localization.
Home Page: https://arxiv.org/abs/1810.10529
where i can get this file. There is nothig like train.txt in your repo. Please help me to
To read the training data run:
ReadList('train.txt', 'Trainset', isTrainSet=True), where train.txt is in the following format: path_to_jpg_image class_label
Hi, Can you share your pretrained model for experimentation?
Hi, can you share your training loss variation? It seems my training loss is not normal...
Epoch (1/20), batch (0/4495), 0.845s, batch err: 5.754796981811523
Epoch (1/20), batch (100/4495), 0.819s, batch err: 19.57297134399414
Epoch (1/20), batch (200/4495), 0.827s, batch err: 1.2522774934768677
Epoch (1/20), batch (300/4495), 0.823s, batch err: 1.128464937210083
Epoch (1/20), batch (400/4495), 0.822s, batch err: 1.730790376663208
Epoch (1/20), batch (500/4495), 0.826s, batch err: 1.2757847309112549
Epoch (1/20), batch (600/4495), 0.826s, batch err: 1.1355090141296387
Epoch (1/20), batch (700/4495), 0.822s, batch err: 1.235153317451477
Epoch (1/20), batch (800/4495), 0.830s, batch err: 1.315584421157837
Epoch (1/20), batch (900/4495), 0.825s, batch err: 1.122027039527893
Epoch (1/20), batch (1000/4495), 0.821s, batch err: 1.2424519062042236
Epoch (1/20), batch (1100/4495), 0.823s, batch err: 8.125507354736328
Epoch (1/20), batch (1200/4495), 0.823s, batch err: 2.4139723777770996
Epoch (1/20), batch (1300/4495), 0.826s, batch err: 2.4237818717956543
Epoch (1/20), batch (1400/4495), 0.818s, batch err: 1.1923820972442627
Epoch (1/20), batch (1500/4495), 0.833s, batch err: 1.8711756467819214
Epoch (1/20), batch (1600/4495), 0.824s, batch err: 1.2039916515350342
Epoch (1/20), batch (1700/4495), 0.823s, batch err: 1.0933177471160889
Epoch (1/20), batch (1800/4495), 0.821s, batch err: 10.37353515625
Epoch (1/20), batch (1900/4495), 0.841s, batch err: 1.202012538909912
Epoch (1/20), batch (2000/4495), 0.831s, batch err: 2.8420190811157227
Epoch (1/20), batch (2100/4495), 0.832s, batch err: 113.91435241699219
Epoch (1/20), batch (2200/4495), 0.834s, batch err: 21.43947982788086
Epoch (1/20), batch (2300/4495), 0.820s, batch err: 156.9434051513672
Epoch (1/20), batch (2400/4495), 0.836s, batch err: 3.8966124057769775
Epoch (1/20), batch (2500/4495), 0.822s, batch err: 1.2476005554199219
Epoch (1/20), batch (2600/4495), 0.849s, batch err: 2.525386333465576
Epoch (1/20), batch (2700/4495), 0.835s, batch err: 1.8467141389846802
Epoch (1/20), batch (2800/4495), 0.823s, batch err: 1.0108225345611572
Epoch (1/20), batch (2900/4495), 0.822s, batch err: 1.1918048858642578
Epoch (1/20), batch (3000/4495), 0.828s, batch err: 1.3513712882995605
Epoch (1/20), batch (3100/4495), 0.820s, batch err: 17.738948822021484
Epoch (1/20), batch (3200/4495), 0.835s, batch err: 1.3260831832885742
Epoch (1/20), batch (3300/4495), 0.824s, batch err: 1.2585244178771973
Epoch (1/20), batch (3400/4495), 0.841s, batch err: 1.078512191772461
Epoch (1/20), batch (3500/4495), 0.821s, batch err: 128.75778198242188
Epoch (1/20), batch (3600/4495), 0.828s, batch err: 177.14918518066406
Epoch (1/20), batch (3700/4495), 0.821s, batch err: 1.050371766090393
Epoch (1/20), batch (3800/4495), 0.824s, batch err: 3.4665944576263428
Epoch (1/20), batch (3900/4495), 0.847s, batch err: 1.1717772483825684
Epoch (1/20), batch (4000/4495), 0.826s, batch err: 1.45053231716156
Epoch (1/20), batch (4100/4495), 0.825s, batch err: 4.0133376121521
Epoch (1/20), batch (4200/4495), 0.824s, batch err: 25.501022338867188
Epoch (1/20), batch (4300/4495), 0.822s, batch err: 1.2257040739059448
Epoch (1/20), batch (4400/4495), 0.823s, batch err: 1.6352043151855469
Epoch (1/20), val err: 40.76654824756441
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.