haltakov/natural-language-image-search

Search photos on Unsplash using natural language

natural-language-image-search's Issues

ValueError: cannot reshape array of size 924786624 into shape (1981161,512)

Getting this error when running this code in the notebook

import pandas as pd
import numpy as np

Load the photo IDs

photo_ids = pd.read_csv("unsplash-dataset/photo_ids.csv")
photo_ids = list(photo_ids['photo_id'])

Load the features vectors

photo_features = np.load("unsplash-dataset/features.npy")

ValueError: cannot reshape array of size 60817344 into shape (1981161,512)

import pandas as pd
import numpy as np

Load the photo IDs

photo_ids = pd.read_csv("unsplash-dataset/photo_ids.csv")
photo_ids = list(photo_ids['photo_id'])

Load the features vectors

photo_features = np.load("unsplash-dataset/features.npy")

Print some statistics

print(f"Photos loaded: {len(photo_ids)}")

After I ran this cell, the error occurred.

Colab Problems with GDrive

The Colab notebook is unable to download the needed files. Using a browser works though.

what if i just want to search by photo without text,what should i do

what if i just want to search by photo without text,what should i do

OSError: Failed to interpret file PosixPath('unsplash-dataset/lite/features/features.npy') as a pickle

OSError: Failed to interpret file PosixPath('unsplash-dataset/lite/features/features.npy') as a pickle

HTTPError

I meet HTTPError: HTTP Error 502: Bad Gateway in display_photo(photo_id)

How to use it on my own dataset?

Potential performance issue: to_csv regression in pandas below 1.4 version

Issue Description:

Hello.
I have discovered a performance degradation in the .to_csv function of pandas version 1.2.0. And I notice the repository depends on pandas 1.2.0 in requirements.txt. I am not sure whether this performance problem in pandas will affect this repository. I found some discussions on pandas GitHub related to this issue, including #39413 and #44908.
I also found that 03-process-unsplash-dataset.ipynb used the influenced api. There may be more files using the influenced api.

Suggestion

I would recommend considering an upgrade to a different version of pandas >= 1.4 or exploring other solutions to optimize the performance of .to_csv.
Any other workarounds or solutions would be greatly appreciated.
Thank you!

Mirror isnt working.

Permission denied: https://drive.google.com/uc?id=1FdmDEzBQCf3OxqY9SbU-jLfH_yZ6UPSj
Maybe you need to change permission over 'Anyone with the link'?
Permission denied: https://drive.google.com/uc?id=1L7ulhn4VeN-2aOM-fYmljza_TQok-j9F
Maybe you need to change permission over 'Anyone with the link'?
--2021-08-24 18:02:28--  https://transfer.army/api/download/9Z976uW4x_Q/MwBsKNX4
Resolving transfer.army (transfer.army)... 18.185.236.87
Connecting to transfer.army (transfer.army)|18.185.236.87|:443... connected.
HTTP request sent, awaiting response... 404 Not Found
2021-08-24 18:02:29 ERROR 404: Not Found.

--2021-08-24 18:02:29--  https://transfer.army/api/download/3bNN8ysWN4U/T7815nW8
Resolving transfer.army (transfer.army)... 18.185.236.87
Connecting to transfer.army (transfer.army)|18.185.236.87|:443... connected.
HTTP request sent, awaiting response... 404 Not Found
2021-08-24 18:02:31 ERROR 404: Not Found.

Maybe host on dropbox?

Missing photos.tsv000 file

Hi,
I am trying to run the code on my machine. 02-download-unsplash-dataset.ipynb requires photos.tsv000 to download the dataset. However there is no such tsv file in the repositiory. Where can I get this file?

Thanks

Unsplash images missing / missing IDs?

Great work on the colab, thank you!

Are the 404s caused by images no longer being on unsplash, or is this an indexing / ID issue? I've registered, logged in and they definitely seem to be missing.