Try out deep learning models online on Google Colab

Thanks for this fantastic resource. I think this collab on real time voice cloning could also be a nice addition to the list: https://colab.research.google.com/drive/1ts-KRxOuuAHmb-AtjU1XTM_lIFOepAN2

Hi Great work on the openpose google colab notebook. Just for my understanding, if i want to run openpose i guess i have to remove the output folder and recompile the binary file ?

When running the Deepsort Notebook I get:

ModuleNotFoundError: No module named 'YOLO3'

not sure how the system path stuff that enables from YOLO3 import YOLO3 works, but that's probably where the error comes from.

Hi following , your example for open-pose we are trying to implement, a simiar dl-colab for 2d pose detector hrnet . We are having some issues , for implementing it , by splitting the input video into frames and applying the model. But for some reason I dont seem to get it working. The github repository we are following are

https://github.com/lxy5513/hrnet

From initial inspection it looks like the model is called here https://github.com/lxy5513/hrnet/blob/f89986411e47749a9ea7eb13b6f3d0f703d03231/pose_estimation/video.py#L140

But we are unable to move forward on this. https://colab.research.google.com/drive/0B0nSo-sEMGvvamtlbl9ucW1ycVF5OWg0QVc3d0E3dmVIQ3FJ

Tips would be appreciated!

Hi, I wanted to understand why tensorflow 1.x was loaded instead of the current version. In Nvidia's requirements they list tensorboardx . I'm new to these things and so want to understand what's really going on here. Thank you!

Hello and thanks for your extensive repo of examples!

I was trying the real-time voice cloning and bounced into this error:

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. tensorflow 1.15.2 requires gast==0.2.2, but you have gast 0.4.0 which is incompatible. lucid 0.3.10 requires numpy<=1.19, but you have numpy 1.20.3 which is incompatible. yellowbrick 1.3.post1 requires numpy<1.20,>=1.16.0, but you have numpy 1.20.3 which is incompatible. kapre 0.3.6 requires tensorflow>=2.0.0, but you have tensorflow 1.15.2 which is incompatible. google-colab 1.0.0 requires requests~=2.23.0, but you have requests 2.27.1 which is incompatible. datascience 0.10.6 requires folium==0.2.1, but you have folium 0.8.3 which is incompatible. albumentations 0.1.12 requires imgaug<0.2.7,>=0.2.5, but you have imgaug 0.2.9 which is incompatible.

Unfortunately I don't know enough of library versioning to fix it myself if you could give me some guidelines...

Thank you!

Roberto

When I tried to run the notebook for Real Time Voice Cloning, it gave me this error:

--2022-01-10 20:13:16-- https://github.com/blue-fish/Real-Time-Voice-Cloning/releases/download/v1.0/pretrained.zip Resolving github.com (github.com)... 140.82.113.3 Connecting to github.com (github.com)|140.82.113.3|:443... connected. HTTP request sent, awaiting response... 404 Not Found 2022-01-10 20:13:17 ERROR 404: Not Found.

FileNotFoundError Traceback (most recent call last) in () 36 from pathlib import Path 37 ---> 38 encoder.load_model(project_name / Path("encoder/saved_models/pretrained.pt")) 39 synthesizer = Synthesizer(project_name / Path("synthesizer/saved_models/pretrained/pretrained.pt")) 40 vocoder.load_model(project_name / Path("vocoder/saved_models/pretrained/pretrained.pt"))

/content/Real-Time-Voice-Cloning/encoder/inference.py in load_model(weights_fpath, device) 30 _device = torch.device(device) 31 _model = SpeakerEncoder(_device, torch.device("cpu")) ---> 32 checkpoint = torch.load(weights_fpath, _device) 33 _model.load_state_dict(checkpoint["model_state"]) 34 _model.eval()

/usr/local/lib/python3.7/dist-packages/torch/serialization.py in load(f, map_location, pickle_module, **pickle_load_args) 592 pickle_load_args['encoding'] = 'utf-8' 593 --> 594 with _open_file_like(f, 'rb') as opened_file: 595 if _is_zipfile(opened_file): 596 # The zipfile reader is going to advance the current file position.

/usr/local/lib/python3.7/dist-packages/torch/serialization.py in _open_file_like(name_or_buffer, mode) 228 def _open_file_like(name_or_buffer, mode): 229 if _is_path(name_or_buffer): --> 230 return _open_file(name_or_buffer, mode) 231 else: 232 if 'w' in mode:

/usr/local/lib/python3.7/dist-packages/torch/serialization.py in init(self, name, mode) 209 class _open_file(_opener): 210 def init(self, name, mode): --> 211 super(_open_file, self).init(open(name, mode)) 212 213 def exit(self, *args):

FileNotFoundError: [Errno 2] No such file or directory: 'Real-Time-Voice-Cloning/encoder/saved_models/pretrained.pt'

I am assuming this is because the pretrained model is no longer up. So I am assuming the fix would just be to change the link to the pretrained model.

ERROR: tensorflow 1.15.2 has requirement gast==0.2.2, but you'll have gast 0.3.3 which is incompatible. ERROR: datascience 0.10.6 has requirement folium==0.2.1, but you'll have folium 0.8.3 which is incompatible. ERROR: albumentations 0.1.12 has requirement imgaug<0.2.7,>=0.2.5, but you'll have imgaug 0.2.9 which is incompatible.

Hi! When running the notebook 'AlphaPhose v0.3.0' as it is, at performing detection on the cut 'video.avi', the line:

!cd {project_name} && python3 scripts/demo_inference.py --sp --video ../video.avi --outdir ../ --save_video --checkpoint ../{pretrained_model_path} --cfg ../{pretrained_model_config_path}

I get the error:

File "scripts/demo_inference.py", line 12, in <module>
    from detector.apis import get_detector
ModuleNotFoundError: No module named 'detector'

So, it tries to run 'demo_inference.py', but can't do the imports. Could you please clarify what could be the reason for this error? Should I add some paths to the working directory? (* the other notebook 'AlphaPose.ipynb' runs fine)

Running the 5th line from LipGAN notebook returns the following error:

#@title  
!rm -rf *.avi  
!cd {project_name} && python batch_inference.py --checkpoint_path logs/lipgan_residual_mel.h5 --face ../{image_name} --fps 30 --audio ../{audio_name} --results_dir ..

2020-04-24 14:55:27.562365: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 Traceback (most recent call last): File "batch_inference.py", line 3, in import scipy, cv2, os, sys, argparse, audio File "/content/LipGAN/audio.py", line 7, in from audio_hparams import hparams as hp File "/content/LipGAN/audio_hparams.py", line 1, in from tensorflow.contrib.training import HParams ModuleNotFoundError: No module named 'tensorflow.contrib'

Hello! I would like to suggest the addition of these two notebooks from works submitted to CPVR2020:

• Learning to See Through Obstructions: remove reflections, fences and water drops from photos https://colab.research.google.com/drive/1iOKknc0dePekUH2TEh28EhcRPCS1mgwz

• 3D Photography using Context-aware Layered Depth Inpainting: create 3D photography from photos https://colab.research.google.com/drive/1706ToQrkIZshRSJSHvZ1RuCiM__YX3Bz

Thanks for the list and have a happy Easter! :)

There's something weird going on, the new links to the pretrained models work fine when executing manually, but when I run the script and check the folder saved_models it is empty.

---------------------------------------------------------------------------
FileNotFoundError                         Traceback (most recent call last)
[<ipython-input-24-4dd965191466>](https://localhost:8080/#) in <module>
     42 
     43 get_ipython().system('ls ')
---> 44 encoder.load_model(project_name / Path("saved_models/default/encoder.pt"))
     45 synthesizer = Synthesizer(project_name / Path("saved_models/default/synthesizer.pt"))
     46 vocoder.load_model(project_name / Path("saved_models/default/vocoder.pt"))

3 frames
[/usr/local/lib/python3.7/dist-packages/torch/serialization.py](https://localhost:8080/#) in __init__(self, name, mode)
    209 class _open_file(_opener):
    210     def __init__(self, name, mode):
--> 211         super(_open_file, self).__init__(open(name, mode))
    212 
    213     def __exit__(self, *args):
FileNotFoundError: [Errno 2] No such file or directory: 'Real-Time-Voice-Cloning/saved_models/default/encoder.pt'

KeyError: "Could not find converter from <class 'densepose.structures.cse.DensePoseEmbeddingPredictorOutput'> to <class 'densepose.structures.chart_result.DensePoseChartResultWithConfidences'>"

Hey @tugstugi

Great job on the repo/collection! :)

The reason for my PR is that I wanted to submit my colab to your repo. I think it will make a nice addition to your list.

Feel free to merge if you like.

Otherwise, thanks again for the great DL colabs.

Alex

Hi! I was trying to use your colab notebook for Pose Detection with OpenPose. Great job, it really helped me :) But unfortunately, every time when I have to use it, I have to install and build OpenPose so I start wondering if there is a way to save the installation. I tried to build it on a folder in my drive but this error occurs:

[ 97%] Linking CXX executable cpp_classification/classification ln: failed to create symbolic link '/content/gdrive/My Drive/Colab Notebooks/OpenPose/openpose/build/caffe/src/openpose_lib-build/examples/cpp_classification/classification.bin': Operation not supported examples/CMakeFiles/classification.dir/build.make:119: recipe for target 'examples/cpp_classification/classification' failed make[5]: *** [examples/cpp_classification/classification] Error 1 make[5]: *** Deleting file 'examples/cpp_classification/classification' CMakeFiles/Makefile2:720: recipe for target 'examples/CMakeFiles/classification.dir/all' failed make[4]: *** [examples/CMakeFiles/classification.dir/all] Error 2