Comments (10)
i just want to konw the accuracy of your ESIM model? Have you got 88%?
This code can't reach 88%, there are some problems with this version code (the mask of attention weight [Model.py, line 160-170] and the mask of mean and max [Model.py, line 220-225]), but even though I fix these bugs, it only got accuracy 84.+% (maybe partly due to different hyperparameters), so I didn't update my code in this repo.
from esim.
lt is hard to got 88%, l didn't got accuracy above 75%, is lt the model implement problems?
from esim.
i just want to konw the accuracy of your ESIM model? Have you got 88%?
This code can't reach 88%, there are some problems with this version code (the mask of attention weight [Model.py, line 160-170] and the mask of mean and max [Model.py, line 220-225]), but even though I fix these bugs, it only got accuracy 84.+% (maybe partly due to different hyperparameters), so I didn't update my code in this repo.
how to fix the bugs above, updated code will be appreciated,thanks
from esim.
i just want to konw the accuracy of your ESIM model? Have you got 88%?
This code can't reach 88%, there are some problems with this version code (the mask of attention weight [Model.py, line 160-170] and the mask of mean and max [Model.py, line 220-225]), but even though I fix these bugs, it only got accuracy 84.+% (maybe partly due to different hyperparameters), so I didn't update my code in this repo.
how to fix the bugs above, updated code will be appreciated,thanks
Ok, I will update it recently. The code on the server have been modified for another research task, I will change it back and test the results in my winter vacation.
from esim.
By adding the mask op, I got the best acc as follow:
2019-01-16 15:01:00 Epoch:23; train: acc:0.9003, macro-f1:0.9003;
2019-01-16 15:01:11 Epoch:23; dev: acc:0.8659, macro-f1:0.8657;
2019-01-16 15:01:21 Epoch:23; test: acc:0.8649, macro-f1:0.8647;
The main hyperparameters is set likes follow:
self.embedding_normalize = 1
self.with_length_mask = True
self.pad_max_length = pad_len # 25
self.hidden_size = 200 # For BiLSTM hidden_size, feedforward hidden_size;
self.learning_rate = 0.05
self.clip_value = None
self.l2_lambda = 0.0
And then I will run the code I fixed, on your paper's hyperparameters.
from esim.
By adding the mask op, I got the best acc as follow:
2019-01-16 15:01:00 Epoch:23; train: acc:0.9003, macro-f1:0.9003;
2019-01-16 15:01:11 Epoch:23; dev: acc:0.8659, macro-f1:0.8657;
2019-01-16 15:01:21 Epoch:23; test: acc:0.8649, macro-f1:0.8647;The main hyperparameters is set likes follow:
self.embedding_normalize = 1
self.with_length_mask = True
self.pad_max_length = pad_len # 25
self.hidden_size = 200 # For BiLSTM hidden_size, feedforward hidden_size;
self.learning_rate = 0.05
self.clip_value = None
self.l2_lambda = 0.0
And then I will run the code I fixed, on your paper's hyperparameters.
hi,i adding the mask op and i trained the model by the hpyerparemeters provided by your post above , but i didnt gei the results.
i got the test acc 68.89 and f1 68.91.
i wanna know how u eidt this code, can i have a look of ur code
my email is [email protected]
look forward ur reply
from esim.
By adding the mask op, I got the best acc as follow:
2019-01-16 15:01:00 Epoch:23; train: acc:0.9003, macro-f1:0.9003;
2019-01-16 15:01:11 Epoch:23; dev: acc:0.8659, macro-f1:0.8657;
2019-01-16 15:01:21 Epoch:23; test: acc:0.8649, macro-f1:0.8647;
The main hyperparameters is set likes follow:
self.embedding_normalize = 1
self.with_length_mask = True
self.pad_max_length = pad_len # 25
self.hidden_size = 200 # For BiLSTM hidden_size, feedforward hidden_size;
self.learning_rate = 0.05
self.clip_value = None
self.l2_lambda = 0.0
And then I will run the code I fixed, on your paper's hyperparameters.hi,i adding the mask op and i trained the model by the hpyerparemeters provided by your post above , but i didnt gei the results.
i got the test acc 68.89 and f1 68.91.
i wanna know how u eidt this code, can i have a look of ur code
my email is [email protected]
look forward ur reply
The code has been sent to your email, pay attention to check it.
from esim.
By adding the mask op, I got the best acc as follow:
2019-01-16 15:01:00 Epoch:23; train: acc:0.9003, macro-f1:0.9003;
2019-01-16 15:01:11 Epoch:23; dev: acc:0.8659, macro-f1:0.8657;
2019-01-16 15:01:21 Epoch:23; test: acc:0.8649, macro-f1:0.8647;
The main hyperparameters is set likes follow:
self.embedding_normalize = 1
self.with_length_mask = True
self.pad_max_length = pad_len # 25
self.hidden_size = 200 # For BiLSTM hidden_size, feedforward hidden_size;
self.learning_rate = 0.05
self.clip_value = None
self.l2_lambda = 0.0
And then I will run the code I fixed, on your paper's hyperparameters.
hi, I'm a beginner of nlp, and I trained the model by my hyperparameters but it can't get great results, test acc only have 65%. So I also want to know how you fix the code, can I have a look of your code too.
my email is [email protected].
look forward to your reply, thank you!
from esim.
By adding the mask op, I got the best acc as follow:
2019-01-16 15:01:00 Epoch:23; train: acc:0.9003, macro-f1:0.9003;
2019-01-16 15:01:11 Epoch:23; dev: acc:0.8659, macro-f1:0.8657;
2019-01-16 15:01:21 Epoch:23; test: acc:0.8649, macro-f1:0.8647;
The main hyperparameters is set likes follow:
self.embedding_normalize = 1
self.with_length_mask = True
self.pad_max_length = pad_len # 25
self.hidden_size = 200 # For BiLSTM hidden_size, feedforward hidden_size;
self.learning_rate = 0.05
self.clip_value = None
self.l2_lambda = 0.0
And then I will run the code I fixed, on your paper's hyperparameters.hi,i adding the mask op and i trained the model by the hpyerparemeters provided by your post above , but i didnt gei the results.
i got the test acc 68.89 and f1 68.91.
i wanna know how u eidt this code, can i have a look of ur code
my email is [email protected]
look forward ur replyThe code has been sent to your email, pay attention to check it.
I also want to know how you fix the code.
my email is [email protected] . thank you
from esim.
I also want to know how you fix the code. i try the code, but ........
my email is [email protected] . thank you
from esim.
Related Issues (10)
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 esim.