profile
viewpoint
Vladimir Iashin v-iashin Tampere, Finland v-iashin.github.io Deep learning engineer, tech enthusiast, Ph.D. student

v-iashin/BMT 46

Source code for "Bi-modal Transformer for Dense Video Captioning" (BMVC 2020)

v-iashin/MDVC 40

PyTorch implementation of Multi-modal Dense Video Captioning (CVPR 2020 Workshops)

v-iashin/CS231n 33

My Solutions for Stanford's CS231n: "CNNs for Visual Recognition"

v-iashin/video_features 30

Extract RAFT and PWC flow frames as well as I3D, R(2+1)D, VGGish, ResNet features from raw videos using multiple GPUs in parallel

v-iashin/VoxCeleb 11

An attempt to replicate the results of [1706.08612] VoxCeleb: a large-scale speaker identification dataset

v-iashin/WebsiteYOLO 9

The back-end for YOLOv3 object detector running online on my website

schatt89/VehicleRecognition 7

Solution for the TAU Vehicle Type Recognition Competition on Kaggle

v-iashin/FoursquareAPI 5

A simple example of using Foursquare API to get the data about a venue (Tips, UserId, and other info) and the data about a user (homecity, gender, number of friends, lists, checkins, photos, and tips) on Python

v-iashin/JumpMethod 4

Selecting a Proper Number of Clusters Using Jumps Method

v-iashin/LearnablePINs 3

An attempt to replicate the results of [1805.00833] Learnable PINs: Cross-Modal Embeddings for Person Identity

push eventv-iashin/CORSMAL-pyAudioAnalysis

Vladimir Iashin

commit sha fbdad776728d349ca1e01766cf12c734ebcdb5f7

added support for private test

view details

push time in 2 hours

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha eac9756114e7df0c625f724ba19476ee7cc0785c

videa_feats: uses all gpus available to the docker

view details

push time in 7 hours

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha a9cd0c9fa88a26400b6b3d6acd0845c771625c3e

(3/3) capacity: rm imges; vid_feats: several gpus

view details

push time in 7 hours

push eventTUNI-ITC/wiki

Vladimir Iashin

commit sha 0b45ad321e05029375d314f3a4483c41d24cbb9c

logo; favicon; email fixes

view details

push time in a day

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha 9519b175e8bc6057f880f9d51465770c68d86658

2/3: capacity: rm calib now takes from the dataset

view details

push time in a day

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha 3b78410459eaade09a893ad7ba9f75c21ab5b760

capacity: attempt private, sorted filelist

view details

push time in a day

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha b5072b86054e5ee8fd949d2c3e9b1d1b68f9501b

vggish adapted for private test

view details

push time in 2 days

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha 622eee7123836008a4c8d288921faca60ea1c66e

a follow-up

view details

push time in 3 days

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha ce05a8de1b302d0150008e298ee201010536b071

dokerfile with env; init main script; feature extr

view details

push time in 3 days

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha 932f57a8182574b58c05a6790a6fd13f34224163

feat extractor scripts

view details

push time in 3 days

push eventv-iashin/CORSMAL

Vladimir Iashin

commit sha c200e7036e58db50a24b7feadc6b433408d807ec

remove features; filepaths and preds

view details

push time in 3 days

create barnchv-iashin/CORSMAL

branch : docker

created branch time in 3 days

push eventTUNI-ITC/wiki

Vladimir Iashin

commit sha aee6ab018b461be3fd26a23476d9aaeec6ef3f7e

index: fix the link to how to contribute

view details

push time in 4 days

push eventTUNI-ITC/wiki

Vladimir Iashin

commit sha 221aa0149a3e42da40859ec9e3388cf36c1561cf

init commit style: color, admonitions support; content: added how to setup remote access index: init commit (links + description README: how to contribute to the wiki

view details

push time in 4 days

push eventTUNI-ITC/wiki

Vladimir Iashin

commit sha 9c363cd77805596f71ca69f97a6c03898a4f17c1

init commit + actions test

view details

push time in 4 days

create barnchTUNI-ITC/wiki

branch : main

created branch time in 4 days

created repositoryTUNI-ITC/wiki

created time in 4 days

PR closed v-iashin/WebsiteYOLO

Updated download_weights_yolov3.sh digitalocean hacktoberfest spam

write a more attractive one i think you are very smart to do so.

+2 -2

0 comment

1 changed file

Chandan-squad

pr closed time in 5 days

issue commentandyljones/coolgpus

Partially-headed servers

Also, if I understand the suggestion of adding Option "AllowEmptyInitialConfiguration" "true" correctly, I can confirm that it didn't work in my case.

More specifically, I added Option "AllowEmptyInitialConfiguration" "true" after this line inside of Driver "nvidia" context https://github.com/andyljones/coolgpus/blob/7a25cf3a0b789397711ace5626eefafcf9193560/coolgpus#L72

andyljones

comment created time in 5 days

issue commentandyljones/coolgpus

console blanks when running coolgpus

Hi

We have a similar setup as you describe and we also face the same issue. I think this is tracked in https://github.com/andyljones/coolgpus/issues/1 and still unsolved.

eugeneware

comment created time in 5 days

push eventv-iashin/v-iashin.github.io

Vladimir Iashin

commit sha 138068ead1c0108a95fbd1ee7b976b3d972b61f2

updated resume

view details

push time in 6 days

issue commentyjxiong/anet2016-cuhk

Dockerfile request

If anyone will come across this issue and have the same request, you can check out https://github.com/yjxiong/anet2016-cuhk/issues/38

lvaleriu

comment created time in 9 days

issue openedyjxiong/anet2016-cuhk

Installation issues: I adapted the TSN Docker container – feel free to use it

Hi everyone!

I was trying to make the code running on my machine but I faced significant trouble installing the packages on my machine as they became obsolete. Then I tried docker. The experience with it wasn't smooth neither because dense_flow installation wasn't compatible with my GPU. So, I tried the docker container provided for the TSN network. I tested it with 1080Ti (worked) and 2080Ti (didn't work w/ CUDNN_STATUS_EXECUTION_FAILED) with 64gb RAM.

docker pull iashin/anet2016-cuhk:latest

or build on top of it by specifying the header of your Dokerfile:

FROM iashin/anet2016-cuhk:latest

The output from the examples:

  • python examples/classify_video.py --use_flow data/plastering.avi
frame sample 53: 0.322223 second
frame sample 54: 0.322393 second
total time: 20.200887 second
----------------Classification Results----------------------
Plastering 0.9559713
Rock climbing 0.020194435
Painting fence 0.011123132
Hanging wallpaper 0.010004199
Painting furniture 0.0007557427
Hand washing clothes 0.00065550825
Laying tile 0.00040866304
Throwing darts 0.00033438313
Paintball 0.00019233384
Hitting a pinata 0.000125195
  • python examples/classify_video.py --use_flow https://www.youtube.com/watch?v=QkuC0lvMAX0
total time: 91.161525 second
----------------Classification Results----------------------
Windsurfing 1.0
Sailing 6.5195724e-16
Surfing 4.3629453e-17
Kite flying 2.1080195e-18
Wakeboarding 5.306861e-19
Snowboarding 6.466279e-21
Plataform diving 3.3573223e-22
Doing motocross 1.9524087e-22
Scuba diving 1.3819956e-22
Fixing bicycle 1.2276588e-22

Here is the Dockerfile I used to build it (also might be useful if you are not familiar with docker and would like to install it on your machine or as a reference for package version)

FROM bitxiong/tsn:cuda9_cudnn7

RUN git clone --recursive https://github.com/yjxiong/anet2016-cuhk
WORKDIR /app/anet2016-cuhk
RUN bash models/get_reference_models.sh

RUN cp -r ../lib/ .
RUN cp ../cv2.so .

RUN pip install easydict==1.6
# YouTube API changes often. Just install the more recent version of youtube_dl by removing the version specification:
RUN pip install youtube_dl==2020.9.20

RUN echo "export ANET_HOME="/app/anet2016-cuhk"" >> ~/.bashrc

created time in 10 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha b141308393cfcaa387426247646831fa8d680965

added warning if saved features are empty

view details

push time in 14 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha e489ec453639c5fb63d7c3c666dd3af8d94f8aab

pwc env: fixed the name of the env

view details

push time in 15 days

issue commentv-iashin/BMT

Multilingual Audio

Yep, there isn't since I regard video_features and BMT as separate projects. Also, you may check how train and val files are formed. I would start by trying to replace the start end fields with something gibberish in the JSON files and see if it throws an error on the way.

I am afraid I cannot provide support for your project because it is related to video captioning but this work is about dense video captioning. Please close the issue if you don't have any more questions about the topic of this issue.

amil-rp-work

comment created time in 15 days

issue closedv-iashin/BMT

RuntimeError: storage has wrong size

I got this in the prediction section (bmt) root@ABC001:/data/zuyuehan/project/BMT# python ./sample/single_video_prediction.py --prop_generator_model_path ./sample/best_prop_model.pt --pretrained_cap_model_path ./sample/best_cap_model.pt --vggish_features_path ./test/pandemic_vggish.npy --rgb_features_path ./test/pandemic_rgb.npy --flow_features_path ./test/pandemic_flow.npy --duration_in_secs 99 --device_id 0 --max_prop_per_vid 100 --nms_tiou_thresh 0.4 Contructing caption_iterator for "train" phase Using vanilla Generator initialization: xavier Glove emb of the same size as d_model_caps Traceback (most recent call last): File "./sample/single_video_prediction.py", line 281, in <module> args.device_id, args.prop_generator_model_path, args.pretrained_cap_model_path, args.max_prop_per_vid File "./sample/single_video_prediction.py", line 94, in load_prop_model checkpoint = torch.load(prop_generator_model_path, map_location='cpu') File "/root/anaconda3/envs/bmt/lib/python3.7/site-packages/torch/serialization.py", line 386, in load return _load(f, map_location, pickle_module, **pickle_load_args) File "/root/anaconda3/envs/bmt/lib/python3.7/site-packages/torch/serialization.py", line 580, in _load deserialized_objects[key]._set_from_file(f, offset, f_should_read_directly) RuntimeError: storage has wrong size: expected 0 got 512 I need your help, and I'd appreciate it if you could help me.

closed time in 15 days

Johnzu-2019

issue commentv-iashin/BMT

RuntimeError: storage has wrong size

No problem. Good luck with your endeavors!

Closing this.

Johnzu-2019

comment created time in 15 days

push eventv-iashin/TuniSurvivalKit

v-iashin

commit sha d24f94494690eb35bb16d05d26972cf1bd984320

remote: tut vpn is no loger supported; setup: fixes

view details

push time in 15 days

push eventv-iashin/BMT

Vladimir Iashin

commit sha 193faaae087293fa220c2a2422d1e8fdcfd6c037

a small readme clarification

view details

push time in 16 days

issue commentv-iashin/BMT

RuntimeError: storage has wrong size

I am afraid I cannot reproduce your error on my side.

I would suggest checking the md5 sums of the models you are trying to load:

md5sum ./sample/best_prop_model.pt
5f8b20826b09eadd41b7a5be662c198b  ./sample/best_prop_model.pt

md5sum ./sample/best_cap_model.pt
7b4d48cd77ec49a027a4a1abc6867ee7  ./sample/best_cap_model.pt

Are you using distributed training of some sort?

Johnzu-2019

comment created time in 16 days

issue commentv-iashin/BMT

OSError: libtorch_cpu.so: cannot open shared object file: No such file or directory

yes i tried to reinstall i3d envs. And i got this error.....

python main.py --feature_type i3d --on_extraction save_numpy --device_ids 0 --extraction_fps 25 --video_paths ../../sample/women_long_jump.mp4 --output_path ../../sample/

ImportError: TensorBoard logging requires TensorBoard with Python summary writer installed. This should be available in 1.14 or above.

Faced the same error when forgot to cd ./submodules/video_features before running the script.

Rubatoo

comment created time in 16 days

issue commentv-iashin/BMT

RuntimeError: storage has wrong size

Hi,

Thanks for reporting. Does it work with the example video from README?

Johnzu-2019

comment created time in 16 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha 25bb7046db2e0c9d804833ef844d7a306898f934

resnet50: support for different fps

view details

push time in 16 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha 02fae36eb78f6b456c23fe9de9fa4d30b083bf52

PWC-Net separate support; i3d: raft support

view details

push time in 16 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha fff275c51600403cef0349ad5d02f17f7f207217

raft: returns only flow_up & default args for test

view details

push time in 17 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha 670289dd80bf6f44ea94fefb8b63db21c5dd57c8

i3d: changed back to cv2 to match raft procedure *added timestamps and fps features* *updated opencv requirement (corrected time stamps)*

view details

push time in 17 days

issue commentv-iashin/BMT

Multilingual Audio

I presume lowering it to say 30 will skip alternate frames. Yes, you can see how ffmpeg does it by:

ffmpeg -i {video_path} -filter:v fps=fps={extraction_fps} {new_path}'

It will save the video with extraction_fps into new_path. Open the video and see it yourself.

How did you come about the conclusion it's idle Well, because V100 > 2080Ti and in my case the load is not 100% and it executes in 7 seconds while yours in 60. Therefore, I just made a guess.

How will increasing the number of GPU's reduce the idle time? There are two parts of the process: extracting/loading frames (done on CPU) and feature extraction (done on GPU). If you have two GPUs you might end up in situations where one GPU is idle and CPU is working (extracts frames/loading them) and the second one is working but the CPU is partially loaded (feature extraction). So, they do different parts you might benefit from having two GPUs. However, reusing the difference between v100 and p100 to invest in a better CPU might actually be a better idea.

You just need to profile your set up a bit and see what takes most of the time. On the video_features master I have an updated I3D such that the reencoded video is saved as an mp4 with diff fps instead of lots of files which might be faster in your case. I do not guarantee the same features as there would be with the BMT state but since you are working on something else rather than a replication of BMT I think you could try these instead.

What impact is created by reducing/increasing step & stack size? Usually, reducing these will increase speed and vice-versa. Since you have ~60sec videos I would go with default parameters (25 fps instead of 60).

amil-rp-work

comment created time in 17 days

issue commentv-iashin/BMT

Multilingual Audio

In my case, it takes 7 seconds (32 threads, 2080ti).

Yep, I think the bottleneck is the CPU. Overall, if you are renting the machine save some $$ on GPU and add more cores and more GPUs (2xP100 > 1xV100) – check the gpu load (I think it is mostly idle).

Also, try to have the temp directory (with frames) on a fast I/O disk (hdd < ssd < nvme) – or even the whole video base.

Anyway, the resolution is not a problem, but the fps is. You will cut the time in half just by specifying --extraction_fps 25 this is the default fps for the original I3D model. So, I would recommend using it.

Let me know how it goes.

As a side note, the path was wrong (thanks for the catch):

--pwc_path ./models/i3d/checkpoints/network-default.pytorch
amil-rp-work

comment created time in 18 days

issue commentv-iashin/BMT

Multilingual Audio

Hm. Well, it could be slow but not this slow.

Can you try to include prints into the code and see what takes most of the time? What are the number of CPU cores and the load?

How long does it take to run examples from the repo (video features)? For instance:

cd video_features
# make sure to be on a specific commit (4fa02bd5c5b8c34081dcfb609e2bcd5a973eaab2)
python main.py --feature_type i3d --device_ids 0 --extraction_fps 25 --stack_size 24 --step_size 24 --pwc_path ./i3d/checkpoints/network-default.pytorch --video_paths ./sample/v_ZNVhz7ctTq0.mp4

How long does it take?

amil-rp-work

comment created time in 18 days

IssuesEvent

issue commentv-iashin/BMT

Multilingual Audio

Hi,

Great!

  • Since we train the caption module first and, then, train the proposal module I would start by just ignoring the proposal part here: https://github.com/v-iashin/BMT#train. Let me know if something will prevent you from doing so. By the way, make sure to extract features first if you haven't yet as the training is done on top of the pre-extracted features. If the I3D will not work for you, you may also try to extract other features (see https://github.com/v-iashin/video_features).
  • I haven't tried to train it on Google Collab. I was using 2080Ti (11gb) with f32 so I think there should not be any problem with P100 and later gpus.
amil-rp-work

comment created time in 18 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha 86a8358e41a123961f2a00edb385036e4266582c

i3d: flow_src -> pwc_src; added raft to readme

view details

Vladimir Iashin

commit sha 6c02da40bf8d4cb1cadc845503de6e2852c24681

i3d: ind streams support, refactoring

view details

Vladimir Iashin

commit sha 144b0d5aa106bc2b104c67edbbd290965bab11f3

i3d: cf and simplified the code; global: imp print

view details

push time in 19 days

create barnchv-iashin/video_features

branch : mdvc

created branch time in 19 days

create barnchv-iashin/video_features

branch : bmt

created branch time in 19 days

issue closedv-iashin/MDVC

ASR

Hello! Then I would like to ask, how did you get your .srt subtitle file?Is the .srt subtitle file downloaded directly on YouTube, or is it obtained by calling Google’s API?

closed time in 21 days

duola1970

issue commentv-iashin/MDVC

ASR

I used Google2SRT

duola1970

comment created time in 21 days

issue commentv-iashin/MDVC

ASR

Hi, I can say that we obtained .srt for each available video and it wasn't through the Google API because the access to it is limited.

duola1970

comment created time in 22 days

push eventv-iashin/BMT

Vladimir Iashin

commit sha 624aac368aad536955662046323380503e298fa9

Make sure to checkout 4fa02bd5c in video_features!

view details

push time in 22 days

push eventv-iashin/MDVC

Vladimir Iashin

commit sha df3b88a8bc10271e9501be41cd77e74d13abf79b

Make sure to checkout 4fa02bd5c in video_features!

view details

push time in 22 days

issue closedv-iashin/MDVC

ASR

Hello! I want to ask what API is used to obtain .srt subtitles?

closed time in 22 days

duola1970

issue commentv-iashin/MDVC

ASR

See https://github.com/v-iashin/MDVC/issues/11#issuecomment-644612209

duola1970

comment created time in 22 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha d7b218c4f72c0f7e20fff5f99b390075d491f1f8

Unified i3d features to the rest of the library * changed the logic from frame extrac to enc w/ diff fps > read_video * * dropped the support for frame reiteration for the final stack (extract_frames_from_video -> form_slices) * * hence, the resizing change is done in transforms rather than during frame extraction * * added AV (read_video) dependency *

view details

push time in 22 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha 3aa4dc141c3b9884bd00f80f7da5c957b9f8dc92

refactored transforms for i3d; updated i3d env

view details

push time in 22 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha cd6292dad5b2c057a651560f3c0705aace512e4b

move raft checkpoints

view details

push time in 23 days

push eventv-iashin/video_features

Vladimir Iashin

commit sha 064c0b3aa65ca8235b2e891ea0fc1866242e62ff

RAFT support; show_class_pred -> show_pred

view details

push time in 23 days

issue closedv-iashin/BMT

Multilingual Audio

Hey @v-iashin Thanks for open sourcing such an awesome work!!! Kudos to you on this and MDVC. I was wondering since my videos are not of English language but I do require captions in the English language, how do I go about utilizing your work here? Is there a possibility to totally ignore audio features and just image features?

closed time in 25 days

amil-rp-work

issue commentv-iashin/BMT

Multilingual Audio

Nope, let me know if you will find/collect one 🙂

Closing this for now.

amil-rp-work

comment created time in 25 days

PR closed v-iashin/WebsiteYOLO

Improved Docs digitalocean hacktoberfest spam
+2 -2

0 comment

1 changed file

ThakurTulsi

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

only added a short form digitalocean hacktoberfest spam

sir i have to do a task in which i have to create 4 pull requests please allow it as it will not make any change in your project please sir

+1 -1

3 comments

1 changed file

sanskar783

pr closed time in a month

pull request commentv-iashin/WebsiteYOLO

only added a short form

@AirFusion45 I know I know (check the closed PRs). Don't worry

@sanskar783 I am sorry but I cannot accept your request as I think the whole idea behind it is explotation rather than creation.

sanskar783

comment created time in a month

issue commentv-iashin/BMT

missing ground truth files (./tmp/extracted_targets_for_{phase}.pkl ).

Hi @minmax100

Please share the full error message.

In the meantime, these files are generated when the training is done for the first time here: https://github.com/v-iashin/BMT/blob/3d0ed089da476787a7a0b84aab207a8e036a419e/datasets/proposal_dataset.py#L133

Does this part fails somehow or why they aren't generated in your case?

minmax100

comment created time in a month

PullRequestEvent

PR closed v-iashin/WebsiteYOLO

Improved Docs digitalocean hacktoberfest spam
+2 -2

0 comment

1 changed file

ThakurTulsi

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improved Docs digitalocean hacktoberfest spam
+2 -1

0 comment

1 changed file

rishabh323

pr closed time in a month

push eventv-iashin/WebsiteYOLO

Vladimir Iashin

commit sha 8a52d4a4403c8660c9045dc2e7e6f007de42b3d7

README update; formatting of some src files

view details

push time in a month

issue commentv-iashin/BMT

Multilingual Audio

You should mind the domain shift between TikTok videos and the videos from ActivityNetCaptions. So, I don't think it is related to non-English audio tracks. And, yes, it is not 100 % accurate, nor any other model. With this being said, I have no evidence to believe that without the audio stream it would work any better in your case. At the same time, I would suggest looking for a dataset with captions which is closer to the dataset you want to apply it on.

amil-rp-work

comment created time in a month

PR closed v-iashin/WebsiteYOLO

Update README.md spam
+6 -6

0 comment

1 changed file

abhinav-bit

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Inspired spam

I am Inspired from Your Work Dear... Its Awesome!

+1 -0

0 comment

1 changed file

faizan7800

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

correct a spell spam
+2 -2

0 comment

1 changed file

SaifVeesar

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Some small changes spam
+6 -6

0 comment

1 changed file

rajaramrom

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improve Docs spam
+1 -1

0 comment

1 changed file

rkgupta95

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improve Docs spam
+1 -1

0 comment

1 changed file

Arpus87

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improve Docs spam
+1 -1

0 comment

1 changed file

MrIconic27

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Updated the ReadMe.md spam
+2 -1

0 comment

1 changed file

iamsandeepprasad

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Updated README.md spam
+1 -1

0 comment

1 changed file

ANKUSHSINGH-PAT

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Update README.md spam

Updated Readme

+2 -0

0 comment

1 changed file

Mahaveer173

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

improve docs spam

in README i can change back-end with bach-end

+1 -1

0 comment

1 changed file

mohitboricha

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Update README spam
+2 -2

0 comment

1 changed file

Ayushgreeshu

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

improve heading spam
+2 -2

0 comment

1 changed file

pawan941394

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Update README.md spam

corrected spelling mistakes

+1 -1

0 comment

1 changed file

priyanka-prasad1

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improved Readme spam

I have corrected some spelling errors.

+2 -2

0 comment

1 changed file

vishalvishw10

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

improved docs spam
+1 -1

0 comment

1 changed file

niteshsharma9

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improve Docs spam
+1 -1

0 comment

1 changed file

Banashree19

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

improved.txt spam
+1 -1

0 comment

1 changed file

sayon-islam-23

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Update README.md spam

improved docs

+1 -1

0 comment

1 changed file

sidhu56

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

IMPROVED THE TITLE spam
+1 -1

0 comment

1 changed file

Sid133

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Updated docs spam
+2 -2

0 comment

1 changed file

white-devil123

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

update spam
+7 -4

0 comment

2 changed files

NEFARI0US

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Doc_Update spam
+1 -1

0 comment

1 changed file

khalane1221

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

improve docs spam
+1 -1

0 comment

1 changed file

Codewithzaid

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Update README.md spam
+3 -1

0 comment

1 changed file

dhruvalgupta2003

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Updated coco.names spam
+10 -0

0 comment

1 changed file

vaibhavimekhe

pr closed time in a month

PR closed v-iashin/WebsiteYOLO

Improved Docs spam
+1 -1

0 comment

1 changed file

marginkantilal

pr closed time in a month

more