Commit e88c2210 authored by Hannah MUCKENHIRN's avatar Hannah MUCKENHIRN

Small modification of README (loading pre-trained models).

parent 8f2f3de7
......@@ -103,13 +103,15 @@ The equal error rate and half total error rate were computed with the [bob](http
If you want to use the pre-trained model (the one obtained in the first step "speaker identification on training set"), you can just load the model in `model/speaker_identification_model.bin` as the following:
`require 'nn'`
`model = torch.load("model/speaker_identification_model.bin")`
```lua
require 'nn'
model = torch.load("model/speaker_identification_model.bin")
net = model[1]
params = model[2]
```
`net = model[1]`
`params = model[2]`
`net` will contain the weights of the trained CNN, `params` will contain the parameters of the CNN specified when running the script `trainSpeakerIdentification.lua` (hyper-parameters, learning rate, normalization scheme...).
# Other Remarks
It is worth mentioning that this code loads the whole data in memory before training a CNN. This means that if you want to train a CNN on a larger database, you should investigate how to load the data in batches.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment