SqueezeNet 1.0 and 1.1 models#49
SqueezeNet 1.0 and 1.1 models#49apaszke merged 6 commits intopytorch:masterfrom Maratyszcza:squeezenet
Conversation
|
I successfully trained SqueezeNet 1.0 on ImageNet with the following parameters: |
|
Awesome! Is the PR ready? |
|
There is SqueezeNet 1.1 to train as well |
|
Finished training both SqueezeNet 1.0 and 1.1. Trained for 100 epochs with
After @soumith uploads the models to AWS, I'll add |
|
Great! If you put them somewhere or send them to me, I can upload them to AWS too. |
|
finished uploading them: |
apaszke
left a comment
There was a problem hiding this comment.
Two minor things. Ready to merge after fixing them.
| ) | ||
| self.classifier = nn.Sequential( | ||
| nn.Dropout(p=0.5), | ||
| self.final_conv, |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
| nn.AvgPool2d(13) | ||
| ) | ||
|
|
||
| import math |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
SqueezeNet 1.0 and 1.1 models. I'm still working on getting them converge on ImageNet, but as several people contacted me about SqueezeNet for PyTorch, I think its useful to share it publicly.