Add in explanation of new EOL Re-align files and the rare-species files, plus links back to original catalogs for TreeOfLife-10M and Rare Species prior to the re-align (as used for v3.3).
Everything in the catalog matches to the cargo archive.
The combined manifests can't pull in all the images from that merge (this is where the missing images come in).
Upload archive analysis and data for catalog realign
We gained an additional 630 matches over previous attempt. None of the images with mismatched archive-manifest IDs matched.
Total number of images missing metadata is 30,746.
3,734 of these are in the train_small subset.
Upload archive analysis and data for rare species
- Generalize stats gen in notebook (length of catalog vs fixed num). (60beed57debc5dde0848b416c87258329827eecf)
- v3.3 EDA update, proper length of catalog, updated stats and stats gen. (edfaa06691fadbb49dfaf9a6beb85301e5b5b83e)
- Move v1.0 visuals into separate folder for record. (e64506dc72e14c5252336a5c1d4038feb893e64c)
- Add large visualization notebooks to LFS tracking. (391a60d6d72590634bf6a19e2f3577963fc1a569)
- Add notebook to generate visuals plus paired py file for diff record. (f96af1395bd064421bd4a535d2c2ac59f5d6701f)
- Add static treemap visualizations for v3.3. (594dbfd34443e01304a214457934d2f881a2652b)
- Add new interactive visuals for v3.3. (d5c38474e7191a734f8c25ce2d5abd57bbf1ecb3)
- Update folder and file descriptions for updated visuals. (1a8bcb0343b953fff80fb11b26d597f68f5e9b49)
- Fix taxa range. (e45e1ecbeca3a5adc3f566d3994c0dce01e14d8a)
- Increase font size on treemap for paper (phylum version). (0b6d6f59e18c2488876b87dc1e12bad518252469)
- Need to track pre-webdataset catalog with LFS. (8dca1b6045337337112546548a552e44a0369131)
- Update data files, including pre-webdataset catalog, this is v3.3. (91beb3aab41c0f7af2ddf8459508b06e9231a4ed)
- Updated EDA in notebooks for v3.3, also added analysis of predicted entries prior to webdataset creation, shows 27K more images expectd. (6b5820466dacab211eafad859fb1ba385cc996ae)
- Added explanation of new files. (67e2a4af8580f9d2ac3f1b7a9ea7a0c74cb0ad9b)
- Added count of unique 7-tuples and genus-species pairs. (e00f09d636815a1caaab72cf72cf23482abdd994)
- Updated dataset file (v3.2), now has split column. (0b1771cf9eb778bea98e498b4006c88fd92285c7)
- Updated statistical information on taxonomic labeling. (8d66838d7c372d9d27f4c0cfa766e590352eee9b)
- Analysis of updated datset. (4fb653fb5d86145834f0812edd3313febea0ff6a)
- Clarified some file descriptions and updated links now that check_taxa script has been merged in open_clip repo. (72df81cd4345fab0f2bc573a4eba1be172fba891)
- Updated data file names to reference they are catalogs, changed read in notebooks to match. (a78ce7f4dac628eb1ce3044d7efbcbb9a30c7cd7)
- Update notebook names to match change of data reference as catalog. (5bcd1783dd4a43463910eaf08f39fa69a792457d)
- Updated references to data and notebooks to reflect renaming. (b6bc3138153c3305c58d082596bc1cb9a2938a67)
- Updated to track specific larger CSV. (0436b05150a5287290e044a17d92e0187a35640d)
- Add output of check_taxa script for original dataset and this update. (c0282544cc8d4543936e1901f9f2f91eb21e1a85)
- Add new stats file (updated v1-dev-names.csv) (5d5c450d1f750b05e244caef02b413eb1377a002)
- Add analysis of updated dataset (statistics.csv) and paired py file for clearer diffs. (da0222428f09c128607ff60ee2736593b6d62255)
- Update file descriptions to include newer files and indicate what is still subject to change. (9b9a4884e3a4d88bff5b0f6f0a21ff6736cfc6e5)
- Add stats (average and standard deviation of image label coverage by taxonomic rank. (84a527b5cd99c1d3064887bb2997dc8184e137ef)
- Updated EDA with some more analysis of BIOSCAN and EOL. (04f793483999ae217b1167ba26fb0b6cb9e0e899)
- Run finished (cf83d94912fe0b6307d03ecec60e5d0d1c0f15d8)
- One last check on unknowns, these will now be weeded out in statistics.csv generation. (0c712963b436a0b6644ad3e40a37d328dcfd1879)
