Comments (4)
With default parameters? I Thoguht I tuned them so that this doesn't happen, sorry about that. As the message suggests, lowering the learning rate does it. Set learning_rate to be about half or fifth of what it is now, until it doesn't explode :)
from neuraltalk.
Here is my result on the default setting:
python driver.py
parsed parameters:
{
"grad_clip": 5,
"rnn_relu_encoders": 0,
"dataset": "flickr8k",
"image_encoding_size": 256,
"eval_max_images": -1,
"drop_prob_decoder": 0.5,
"word_encoding_size": 256,
"max_epochs": 50,
"eval_batch_size": 100,
"fappend": "baseline",
"generator": "lstm",
"write_checkpoint_ppl_threshold": -1,
"decay_rate": 0.999,
"tanhC_version": 0,
"hidden_size": 256,
"momentum": 0.0,
"worker_status_output_directory": "status/",
"learning_rate": 0.001,
"checkpoint_output_directory": "cv/",
"do_grad_check": 0,
"word_count_threshold": 5,
"batch_size": 100,
"regc": 1e-08,
"smooth_eps": 1e-08,
"solver": "rmsprop",
"eval_period": 1.0,
"drop_prob_encoder": 0.5
}
Initializing data provider for dataset flickr8k...
BasicDataProvider: reading data/flickr8k/dataset.json
BasicDataProvider: reading data/flickr8k/vgg_feats.mat
preprocessing word counts and creating vocab based on word count threshold 5
253/15000 batch done in 3.242s. at epoch 0.84. loss cost = 39.264201, reg cost = 0.000001, ppl2 = 29.60 (smooth 47.89)
254/15000 batch done in 3.133s. at epoch 0.85. loss cost = 39.633654, reg cost = 0.000001, ppl2 = 33.57 (smooth 47.74)
255/15000 batch done in 3.169s. at epoch 0.85. loss cost = 38.571550, reg cost = 0.000001, ppl2 = 29.56 (smooth 47.56)
.
.
.
.
.
one day later...
.
14999/15000 batch done in 3.492s. at epoch 50.00. loss cost = 28.621228, reg cost = 0.000004, ppl2 = 11.19 (smooth 10.80)
evaluating val performance in batches of 100
evaluated 5000 sentences and got perplexity = 17.785250
validation perplexity = 17.785250
from neuraltalk.
@StevenLOL Nice! Looking at the Model Zoo,
http://cs.stanford.edu/people/karpathy/neuraltalk/
my LSTM model achieves perplexity of about 15.7 (which is slightly better). I ran it for longer and cross-validated it on our cluster, though.
from neuraltalk.
Thanks I will try again with reduced learning rate
On Jan 10, 2015, at 10:59 AM, Steven [email protected] wrote:
Here is my result on the default setting:
python driver.py
parsed parameters:
{
"grad_clip": 5,
"rnn_relu_encoders": 0,
"dataset": "flickr8k",
"image_encoding_size": 256,
"eval_max_images": -1,
"drop_prob_decoder": 0.5,
"word_encoding_size": 256,
"max_epochs": 50,
"eval_batch_size": 100,
"fappend": "baseline",
"generator": "lstm",
"write_checkpoint_ppl_threshold": -1,
"decay_rate": 0.999,
"tanhC_version": 0,
"hidden_size": 256,
"momentum": 0.0,
"worker_status_output_directory": "status/",
"learning_rate": 0.001,
"checkpoint_output_directory": "cv/",
"do_grad_check": 0,
"word_count_threshold": 5,
"batch_size": 100,
"regc": 1e-08,
"smooth_eps": 1e-08,
"solver": "rmsprop",
"eval_period": 1.0,
"drop_prob_encoder": 0.5
}
Initializing data provider for dataset flickr8k...
BasicDataProvider: reading data/flickr8k/dataset.json
BasicDataProvider: reading data/flickr8k/vgg_feats.mat
preprocessing word counts and creating vocab based on word count threshold 5253/15000 batch done in 3.242s. at epoch 0.84. loss cost = 39.264201, reg cost = 0.000001, ppl2 = 29.60 (smooth 47.89)
254/15000 batch done in 3.133s. at epoch 0.85. loss cost = 39.633654, reg cost = 0.000001, ppl2 = 33.57 (smooth 47.74)
255/15000 batch done in 3.169s. at epoch 0.85. loss cost = 38.571550, reg cost = 0.000001, ppl2 = 29.56 (smooth 47.56).
.
.
.
.
one day later...
.14999/15000 batch done in 3.492s. at epoch 50.00. loss cost = 28.621228, reg cost = 0.000004, ppl2 = 11.19 (smooth 10.80)
evaluating val performance in batches of 100
evaluated 5000 sentences and got perplexity = 17.785250
validation perplexity = 17.785250—
Reply to this email directly or view it on GitHub.
from neuraltalk.
Related Issues (20)
- training over new dataset HOT 4
- init_model_from argument has no effect on where driver starts HOT 2
- question about dropout implementation HOT 3
- change in extract_features needed due to caffe update HOT 1
- predict_on_images.py: error: too few arguments HOT 1
- Please someone explain the magic behind this
- Use for sentence input to sentence output
- Confusion about an equation in the paper
- Have you implemented Visual-Semantic Alignments ? HOT 3
- Maybe a mistake in lstm_generator.py HOT 2
- How can i use this code to train regions & snippets RNN model? HOT 3
- when i train a batch(batchsize 100) of flickr8k, it takes around 0.8 seconds, HOT 4
- Incorrect prediction while testing.
- Why optimizing the Ws matrix directly?
- Can we use levelDB features extracted using C++ utility at the place of vgg_feats.mat?
- vgg_feats.mat file is being created but its having very small size of 194bytes. Why so?
- image captioning
- multi-bleu.perl
- eval_sentence_predictions.py: error: too few arguments HOT 3
- predict_on_images.py error
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 neuraltalk.