Add README
Browse files
README.md
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Download datasets:
|
| 2 |
+
* Download and decompress tsv file from here: https://github.com/google-research-datasets/wit/blob/main/DATA.md
|
| 3 |
+
* Use `prepare_wit.py` to download images from Wikipedia.
|
| 4 |
+
* Use `discard_incorrect_files` to filter out corrupt files.`TODO: Still some corrupt files are being kept.` `TODO: Make it a CLI`.
|
| 5 |
+
* Finally, use `run-clip.sh` to train.
|