rd20karim / m2t-segmentation Goto Github PK

Official implementation of the paper Motion2Language, Unsupervised learning of synchronized semantic motion segmentation

License: MIT License

Python 100.00%

attention-mechanism motion-to-text semantic-analysis semantic-segmentation motion-captioning synchronization

m2t-segmentation's People

Contributors

Stargazers

Watchers

Forkers

euromovdhm-semtaxm

m2t-segmentation's Issues

Error in data pre-processing

When I run step 3 to generate sentence_corrections.csv, I get the following error:

    [INFO] CORRECT TOKENS True
    [INFO] INITIAL CORRECTIONS AND LOWER CASING ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, five --> 5 
    [INFO] Replacing ...
    [INFO] Text to int, one --> 1 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 15 corrected perso ----> person 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, one --> 1 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, five --> 5 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 47 corrected  ----> i 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] EMPTY TOKEN [[ ]]
    [INFO] token   not found
    [INFO] Token index 61 corrected foward ----> forward 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, four --> 4 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 69 corrected sim ----> him 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 71 corrected  ----> i 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, one --> 1 
    [INFO] Text to int, three --> 3 
    [INFO] Replacing ...
    [INFO] Text to int, five --> 5 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, two --> 2 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, five --> 5 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 116 corrected counter- ----> counter 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 117 corrected  ----> i 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, three --> 3 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Text to int, two --> 2 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 134 corrected staight ----> straight 
    [INFO] Token index 134 corrected soemthing ----> something 
    [INFO] Token index 136 corrected ’s ----> is 
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Replacing ...
    [INFO] Token index 151 corrected ... ----> None 
    Traceback (most recent call last):
      File "/home/dhegde/projects/motion-pipe/M2T-Segmentation/datasets/kit_m2t_dataset.py", line 203, in <module>
        data = dataset_class(path, filter_data=True,min_freq=3)
      File "/home/dhegde/projects/motion-pipe/M2T-Segmentation/datasets/kit_m2t_dataset.py", line 46, in __init__
        self.lang = vocabulary(self.sentences, correct_tokens=correct_tokens, ask_user=False)
      File "/home/dhegde/projects/motion-pipe/M2T-Segmentation/datasets/vocabulary.py", line 94, in __init__
        self.token_correction(ask_user)
      File "/home/dhegde/projects/motion-pipe/M2T-Segmentation/datasets/vocabulary.py", line 158, in token_correction
        desc = " ".join(tokens)
    TypeError: sequence item 4: expected str instance, NoneType found

Vocab size mismatch

I processed the data as mentioned in the README and get the following error when I run

 python visualizations/poses2concepts.py --config ./configs_kit_aug/MLP.yaml --dataset_name kit --n_map 5 --n_gifs 50 --save_results ./gifs_map_kit22_

ERROR MESSAGE:

    2024-05-01 18:12:11,523 [INFO] Building dataset ... 
    2024-05-01 18:12:11,823 [INFO] CORRECT TOKENS False
    2024-05-01 18:12:12,073 [INFO] Building vocabulary with minimum frequency : 3
    2024-05-01 18:12:12,422 [INFO] Std/Mean Normalization
    2024-05-01 18:12:13,426 [INFO] Convert token to numerical values
    2024-05-01 18:12:13,767 [INFO] Using official split
    2024-05-01 18:12:13,853 [INFO] Number  of samples VAL: 470 TRAIN: 7451, TEST: 1376 
    2024-05-01 18:12:16,133 [INFO] VOCAB SIZE  = 1709 
    2024-05-01 18:12:17,974 [INFO] Scaling 1.0
    2024-05-01 18:12:17,981 [INFO] Applying local_recurrent attention
    Traceback (most recent call last):
      File "/home/dhegde/projects/motion-pipe/M2T-Segmentation/visualizations/poses2concepts.py", line 180, in <module>
        loaded_model, train_data_loader, test_data_loader,val_data_loader = load_model_config(device=device,args=args)
      File "/home/dhegde/projects/motion-pipe/M2T-Segmentation/src/evaluate_m2L.py", line 76, in load_model_config
        loaded_model.load_state_dict(new_dict)
      File "/home/dhegde/anaconda3/envs/llava/lib/python3.10/site-packages/torch/nn/modules/module.py", line 2041, in load_state_dict
        raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
    RuntimeError: Error(s) in loading state_dict for seq2seq:
            size mismatch for dec.embedding.weight: copying a param with shape torch.Size([878, 64]) from checkpoint, the shape in current model is torch.Size([1709, 64]).
            size mismatch for dec.fc.weight: copying a param with shape torch.Size([878, 192]) from checkpoint, the shape in current model is torch.Size([1709, 192]).
            size mismatch for dec.fc.bias: copying a param with shape torch.Size([878]) from checkpoint, the shape in current model is torch.Size([1709]).

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.