Comments (12)
Thank you @nicholasguimaraes i will check this
from action-recognition.
the same problem
from action-recognition.
the same problem
There were several variable issues with the extract_frames.py. I have opened a pull request for that.
#4
from action-recognition.
another error after the modification
RuntimeError: Error(s) in loading state_dict for ConvLSTM:
Missing key(s) in state_dict: "lstm.lstm.weight_ih_l0_reverse", "lstm.lstm.weight_hh_l0_reverse", "lstm.lstm.bias_ih_l0_reverse", "lstm.lstm.bias_hh_l0_reverse", "output_layers.0.weight", "output_layers.0.bias", "output_layers.1.weight", "output_layers.1.bias", "output_layers.1.running_mean", "output_layers.1.running_var", "output_layers.3.weight", "output_layers.3.bias", "attention_layer.weight", "attention_layer.bias".
Unexpected key(s) in state_dict: "lstm.final.0.weight", "lstm.final.0.bias", "lstm.final.1.weight", "lstm.final.1.bias", "lstm.final.1.running_mean", "lstm.final.1.running_var", "lstm.final.1.num_batches_tracked", "lstm.final.3.weight", "lstm.final.3.bias".
from action-recognition.
Had the same issue @WangZhaoxue , did anyone run this project besides the author?
from action-recognition.
Had the same issue @WangZhaoxue , did anyone run this project besides the author?
I have successfully set up the environment and run the code. (Anaconda3-2019.03 python=3.6 pytorch 1.1.0)
There are some typo errors in this repo.
I did some minor fixes here. #4
Also, make sure the extraction process is completed without error.
from action-recognition.
Thank you @WangZhaoxue for your help. I made the minor changes and still have the same error, but I must use PyTorch 1.2 or 1.3 and I think that´s the issue why the loaded can´t load. I try to fix it and response if it´s works.
Missing key(s) in state_dict: "lstm.lstm.weight_ih_l0_reverse", "lstm.lstm.weight_hh_l0_reverse", "lstm.lstm.bias_ih_l0_reverse", "lstm.lstm.bias_hh_l0_reverse", "output_layers.0.weight", "output_layers.0.bias", "output_layers.1.weight", "output_layers.1.bias"....
from action-recognition.
Thank you @WangZhaoxue for your help. I made the minor changes and still have the same error, but I must use PyTorch 1.2 or 1.3 and I think that´s the issue why the loaded can´t load. I try to fix it and response if it´s works.
Missing key(s) in state_dict: "lstm.lstm.weight_ih_l0_reverse", "lstm.lstm.weight_hh_l0_reverse", "lstm.lstm.bias_ih_l0_reverse", "lstm.lstm.bias_hh_l0_reverse", "output_layers.0.weight", "output_layers.0.bias", "output_layers.1.weight", "output_layers.1.bias"....
Check out this issue. They found a solution for this error. It seems that you need to edit the name of the layers and add an attention layer to the pre trained model.
I'll try it out later today and update here if I was able to advance or not.
from action-recognition.
@nicholasguimaraes could you tell the solution for this problem
from action-recognition.
@nicholasguimaraes could you tell the solution for this problem
Hello @sathees07,
I still didn't try testing the pre trained model but if you want to load the model provided by erik you will need to update the names of the layers in the state_dict.
Also, the pretrained model does not have an attention layer so, it seems that you should add that when fixing the state_dict.
Here is a link where you can check how to access the names of the layers of your model through its state_dict.
https://pytorch.org/tutorials/beginner/saving_loading_models.html#what-is-a-state-dict
from action-recognition.
Did you solve this problem?
from action-recognition.
No @js-kang
from action-recognition.
Related Issues (20)
- Hello Mr.Linder-Norén
- list index out of range when start training HOT 1
- Testing performance on official UCF-101 split 1 HOT 5
- AttributeError: 'Namespace' object has no attribute 'sequence_length' HOT 1
- latent_att not defined before. HOT 3
- Inspired from your work, I updated this repo to use fastai2
- Small bug fix
- Terminology mistake
- train.py: error: ambiguous option: --img_dim could match --img_dim_H, --img_dim_W HOT 1
- when I try to use your pretrained model,it give some error.... HOT 4
- size mismatch for output_layers.3.bias: copying a param with shape torch.Size([101]) from checkpoint, the shape in current model is torch.Size([105]). HOT 1
- ValueError: not enough values to unpack (expected 2, got 1)
- how i can solve it HOT 1
- Softmax in Model Output, then using CE Loss HOT 1
- Error: Regarding Loading Pre-Trained Weights HOT 3
- how long it will cost on cpu machine when predict one video.
- Paper link for this repository!!! HOT 1
- Train issue HOT 4
- test 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 action-recognition.