Trouble getting submissions to run

Unfortunately, The logs for the Generate Predictions On Test Data part of the evaluation remain private and only available for admins. It is due to the private test data used in Generate Predictions On Test Data part of the evaluation.

View post 0

Trouble getting submissions to run

I looked into the notebook you submitted in the submission #171169 and the error in our internal logs was IndexError: index 0 is out of bounds for axis 0 with size 0 occured on the line X1.append(df_saved_vocab[df_saved_vocab['word']==t]['min'].values[0]) . I tried executing the notebook locally with the private data and found out the error is due to the list returned by df_saved_vocab[df_saved_vocab['word']==t]['min'].values had no elements ( [] ) and when you tried to get the 0 index element by using .values[0] it resulted in the above error IndexError: index 0 is out of bounds for axis 0 with size 0 occured` .

In Generate Predictions On Test Data phrase of evaluation, we have private test data of over 20k samples used for generating the predictions. In your df_saved_vocab.csv there are values of only 95 different words which can be insufficient given that our private test data has over 20k samples.

What I will suggest to try is -

1. You can either generate the embeddings in the Prediction phase of the notebook
2. You can generate the embeddings of top n most common English words and then save it in the df_saved_vocab.csv.
3. You can also add try except block in the .values[0] to make sure that if this error occurs, you can append, for ex. ( [0] ) to the X1 list.

I hope this explanation helps Let me know if you have any more doubts

Best
Shubhamai

View post 1

Trouble getting submissions to run

Hi mkeywood

I am looking into this issue and will get back to you asap

Best
Shubhamai

View post 0

Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0:

Hi pn1

The warning is due to because there are no GPUs in the evaluator. The evaluator in the NLP Feature Engineering puzzle only uses CPU for evaluation. This warning will not abort the evaluation. I hope this helps

Best,
Shubhamai

View post 0

How to deal with draws

Strange, the 9008.jpg of test doesn’t looks like a draw to me. Can you check if this is the right image.

Also, by illegal, do you only mean that in case of checkmate the player still have turn ? I am checking in my code to figure out the root of problem but can you share some samples of this case.

Shubhamai

View post 0

How to deal with draws

Hi boussif oussama .

I guess you are talking about Win prediction in which case can you please share more examples that have draws in the test set.

I am looking into the second issue and will give updates

Shubhamai

View post 1

About white passed pawn

Nope, it’s not a mistake, we are not differentiating between black/white queen in the samples of this challenge, including test ground truth, so just put small q if you find moves which have conversion of pawn to queen. Enjoy

View post 0

More masks than bboxes?

Hi, thanks for pointing this out! We are looking into this issue and giving updates asap

View post 0

Detectron2 Colab Notebook from Data Exploration to Training the Model 🚀

Hi Guys, Me Shubhamai, Machine Learning Engineer working extensively in the Deep Learning & Computer Vision field.

I have created a colab notebook for this competition which contains a lot of sections including data exploration/visualizations and making model & training it and things like that.

newplot (6)700×450 15.3 KB

newplot (7)700×450 11.5 KB

Here is the link to the colab notebook . I hope that it will help, i would also add more things into that soon.

View post 3

No one sample in new 'test_black_white_images-v2.zip'

(post withdrawn by author, will be automatically deleted in 24 hours unless flagged)

View post 0

[Explainer] Detectron2 & COCO Dataset 🔥 • Web Application & Visualizations • End-to-End Baseline & Tensorflow

So, me Shubhamai and I have come up with these 3 things -

COCO Dataset & using Detectron2, MMDetection

YES! I have converted this dataset into COCO Dataset and which we train Mask-RCNN using Detectron2.

There we go boys - Colab Link

More things will be added so like this post RIGHT NOW

Web Application & Visualisation

https://seismic-facies-identification.herokuapp.com/

But this time, I found that a great preprocessing pipeline can help to model to find accurate features and increasing overall accuracy. But it kinda isn’t that easy as it looks —

So I made a Web Application based on that which allows you to play/experiment with many of the image preprocessing functions/methods, changing parameters or writing custom image preprocessing functions to experiment.

Annotation 2020-09-21 1913421920×992 239 KB

And it also contains all the visualizations from the colab notebook .

I hope that it will help you in making the perfect preprocessing pipelines .

End-to-End Baseline & Tensorflow

https://colab.research.google.com/drive/1t1hF_Vs4xIyLGMw_B9l1G6qzLBxLB5eG?usp=sharing

I have made a complete colab notebook from Data Exploration to Submitting Predictions. Here are some of the glimpse of the image visualization section!

newplot (4)1100×600 168 KB

And this 3D Plot!

newplot (5)1100×600 196 KB

1100×600 196 KB

Tables of Content -

1. Setting our Workspace
2. Data Exploration
3. Image Preprocessing Techniqes
4. Creating our Dataset
5. Creating our Model
6. Training the Model
7. Evaluating the model
8. Testing on test Data
9. Generate More Data + Some tips & tricks

The main libraries covered in this notebook is —

• Tensorflow 2.0 & Keras
• Plotly
• cv2
  and much more…

The model that i am using is UNet, pretty much standard in image segmentation. More is in the colab notebook!

I hope the colab notebook will help you get started in this competition or learning something new . If the notebook did help you, make sure to like the post. lol.

https://colab.research.google.com/drive/1t1hF_Vs4xIyLGMw_B9l1G6qzLBxLB5eG?usp=sharing

Please like the topic if this helps in any way possible . I really appreciate that

View post 3

📝 Explained by the Community | Win 4 x DJI Mavic Drones

Hello Everyone!

Me Shubhamai, Machine Learning Engineer, and I am excitedly working on this competition because especially it a kind of 3D problem rather than 2D. Previously I made the complete google colab notebook from data exploration to submission. You can find the notebook here.

But this time, I found that a great preprocessing pipeline can help to model to find accurate features and increasing overall accuracy. But it kinda isn’t that easy as it looks —

So I made a Web Application based on that which allows you to play/experiment with many of the image preprocessing functions/methods, changing parameters or writing custom image preprocessing functions to experiment.

Annotation 2020-09-21 1913421920×992 239 KB

And it also contains all the visualizations from the colab notebook.

I hope that it will help you in making the perfect preprocessing pipelines and make sure you like the post . Thanks

https://seismic-facies-identification.herokuapp.com/

View post 8

📝 Explained by the Community | Win 4 x DJI Mavic Drones

Sure! Thanks for the suggestion . I will make changes to the color map soon!

View post 0

📝 Explained by the Community | Win 4 x DJI Mavic Drones

Hi Everyone

I have made a complete colab notebook from Data Exploration to Submitting Predictions. Here are some of the glimpse of the image visualization section!

newplot (4)1100×600 168 KB

And this 3D Plot!

newplot (5)1100×600 196 KB

Tables of Content -

1. Setting our Workspace

2. Data Exploration

3. Image Preprocessing Techniqes

4. Creating our Dataset

5. Creating our Model

6. Training the Model

7. Evaluating the model

8. Testing on test Data

9. Generate More Data + Some tips & tricks

The main libraries covered in this notebook is —

• Tensorflow 2.0 & Keras
• Plotly
• cv2
  and much more…

The model that i am using is UNet, pretty much standard in image segmentation. More is in the colab notebook!

I hope the colab notebook will help you get started in this competition or learning something new . If the notebook did help you, make sure to like the post. lol.

https://colab.research.google.com/drive/1t1hF_Vs4xIyLGMw_B9l1G6qzLBxLB5eG?usp=sharing

More things will be added soon!

By

View post 7

Explained by the Community | 100 CHF Prize contest 🎉

Hi,

Me Shubhamai, i did the LNDST competition and here is the github repo https://github.com/Shubhamai/water-segmentation

View post 2