egrace479 commited on
Commit
cbd6949
1 Parent(s): b5dfedd

Adjust links and filepaths according to re-organization in b5dfedde

Browse files
README-supplemental.md CHANGED
@@ -1,7 +1,7 @@
1
  # Summary of Deduplication of Jiggins Data
2
 
3
  We explored our downloaded images in `notebooks/EDA-DL-0-3.ipynb`, and established the following causes of duplication along with remedies:
4
- 1. The "patch" record (record 3477891) which required alignment in `../notebooks/Data-gen-1-1.ipynb`, resulting in `Jiggins_Zenodo_Img_Master_3477891Patch.csv` used for download, is a duplication of all listed records. Additionally, it introduced extra copies of 4 images with incorrect labels for their views (i.e., dorsal vs ventral).
5
  - Solution: This record will be removed.
6
  2. The remaining duplication (430 images x2) comes from 5 records: [4291095](https://zenodo.org/records/4291095), [2813153](https://zenodo.org/records/2813153), [5526257](https://zenodo.org/records/5526257), [2553977](https://zenodo.org/records/2553977), and [2552371](https://zenodo.org/records/2552371).
7
  - Though there is nothing in their descriptions on Zenodo to indicate this overlap occurred, record 4291095 duplicated 415 images from record 2813153. 104 of these images had their view/side labels (dorsal vs ventral) updated to reflect an earlier mislabeling in record 2813153; these were all RAW copies of the images. All other metadata was consistent across records.
@@ -20,7 +20,7 @@ We explored our downloaded images in `notebooks/EDA-DL-0-3.ipynb`, and establish
20
  6. Be sure to check only desired categories are included, such as excluding hybrids or cross types, or specimens not labeled to the level of classification.
21
 
22
 
23
- Final deduplication following this process was completed in `../notebooks/Data-gen-1-2.ipynb`.
24
 
25
  # Other Notes on Using this Data
26
 
 
1
  # Summary of Deduplication of Jiggins Data
2
 
3
  We explored our downloaded images in `notebooks/EDA-DL-0-3.ipynb`, and established the following causes of duplication along with remedies:
4
+ 1. The "patch" record (record 3477891) which required alignment in `notebooks/Data-gen-1-1.ipynb`, resulting in `metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv` used for download, is a duplication of all listed records. Additionally, it introduced extra copies of 4 images with incorrect labels for their views (i.e., dorsal vs ventral).
5
  - Solution: This record will be removed.
6
  2. The remaining duplication (430 images x2) comes from 5 records: [4291095](https://zenodo.org/records/4291095), [2813153](https://zenodo.org/records/2813153), [5526257](https://zenodo.org/records/5526257), [2553977](https://zenodo.org/records/2553977), and [2552371](https://zenodo.org/records/2552371).
7
  - Though there is nothing in their descriptions on Zenodo to indicate this overlap occurred, record 4291095 duplicated 415 images from record 2813153. 104 of these images had their view/side labels (dorsal vs ventral) updated to reflect an earlier mislabeling in record 2813153; these were all RAW copies of the images. All other metadata was consistent across records.
 
20
  6. Be sure to check only desired categories are included, such as excluding hybrids or cross types, or specimens not labeled to the level of classification.
21
 
22
 
23
+ Final deduplication following this process was completed in `notebooks/Data-gen-1-2.ipynb`.
24
 
25
  # Other Notes on Using this Data
26
 
README.md CHANGED
@@ -176,7 +176,7 @@ These images are a subset of the [Butterfly Genetics Group's Cambridge butterfly
176
 
177
  Data is pulled from the Zenodo Records in [`licenses.json`](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/metadata/licenses.json). This file also contains full citation information for all records.
178
 
179
- The [original Master file](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/metadata/Jiggins_Zenodo_Master.csv), created by Christopher Lawrence, was fed into notebook Data-gen-0-1, and further processed in Data-gen-0-2 and Data-gen-0-3. The next data generation step (deduplication by MD5) involved downloading all images in `Jiggins_Zenodo_Img_Master_3477891Patch.csv` (created in Data-gen-1-1). MD5s of all downloaded images were taken and the results were explored in the EDA-DL series of notebooks; information and conclusions from this EDA are documented in the [`deduplication_process` README](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/deduplication_process/README.md). The outline described there was then implemented in the [Data-gen-1-2 notebook](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/notebooks/Data-gen-1-2.ipynb), regenerating all current datasets. This notebook also has another round of taxonomic standardization so that all entries labeled only to the genus level are labeled as `<Genus> sp.` in both `Taxonomic_Name` and `species` columns, 4 entries with a non-butterfly genus name were noted and removed, as were 18 entries with `Unknown` as their taxonomic label.
180
 
181
 
182
  #### Initial Data Collection and Annotation
@@ -199,7 +199,7 @@ None
199
  - There are multiple images of the same specimen for many specimens; sometimes this is due to different views (eg., dorsal or ventral side), sometimes it is due to JPG copies of the RAW photos, though it seems it is also sometimes that new photos were taken of the same specimen at a different times.
200
  - The master files contain only images that were determined to be unique (at the pixel level) through MD5 checksum. This does _**not**_ guarantee that there are not images that are cropped copies of other photos. For instance, [record 2553977](https://zenodo.org/records/2553977) has a number of images with `_cut_` in their name, some of which are close-up crops of the butterfly wings (though not all, some are just close-ups of the full butterfly).
201
 
202
- Please see the [`deduplication_process/README.md`](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/deduplication_process/README.md) for more details on the deduplication process and recommendations on how to split this data effectively without data leakage.
203
 
204
 
205
  ## Additional Information
 
176
 
177
  Data is pulled from the Zenodo Records in [`licenses.json`](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/metadata/licenses.json). This file also contains full citation information for all records.
178
 
179
+ The [original Master file](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/metadata/Jiggins_Zenodo_Master.csv), created by Christopher Lawrence, was fed into notebook Data-gen-0-1, and further processed in Data-gen-0-2 and Data-gen-0-3. The next data generation step (deduplication by MD5) involved downloading all images in `metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv` (created in Data-gen-1-1). MD5s of all downloaded images were taken and the results were explored in the EDA-DL series of notebooks; information and conclusions from this EDA are documented in [`README-supplemental`](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/README-supplemental.md). The outline described there was then implemented in the [Data-gen-1-2 notebook](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/notebooks/Data-gen-1-2.ipynb), regenerating all current datasets. This notebook also has another round of taxonomic standardization so that all entries labeled only to the genus level are labeled as `<Genus> sp.` in both `Taxonomic_Name` and `species` columns, 4 entries with a non-butterfly genus name were noted and removed, as were 18 entries with `Unknown` as their taxonomic label.
180
 
181
 
182
  #### Initial Data Collection and Annotation
 
199
  - There are multiple images of the same specimen for many specimens; sometimes this is due to different views (eg., dorsal or ventral side), sometimes it is due to JPG copies of the RAW photos, though it seems it is also sometimes that new photos were taken of the same specimen at a different times.
200
  - The master files contain only images that were determined to be unique (at the pixel level) through MD5 checksum. This does _**not**_ guarantee that there are not images that are cropped copies of other photos. For instance, [record 2553977](https://zenodo.org/records/2553977) has a number of images with `_cut_` in their name, some of which are close-up crops of the butterfly wings (though not all, some are just close-ups of the full butterfly).
201
 
202
+ Please see [`README-supplemental.md`](https://huggingface.co/datasets/imageomics/Jiggins_Heliconius_Collection/blob/main/README-supplemental.md) for more details on the deduplication process and recommendations on how to split this data effectively without data leakage.
203
 
204
 
205
  ## Additional Information
notebooks/Data-gen-1-1.ipynb CHANGED
@@ -34,7 +34,7 @@
34
  }
35
  ],
36
  "source": [
37
- "multimedia = pd.read_csv(\"../deduplication_process/metadata/Zenodo_meta_files/multimedia__(rec_3477891).csv\", low_memory=False)\n",
38
  "multimedia.info(show_counts=True)"
39
  ]
40
  },
@@ -481,7 +481,7 @@
481
  }
482
  ],
483
  "source": [
484
- "occurrence = pd.read_csv(\"../deduplication_process/metadata/Zenodo_meta_files/occurrences__(rec_3477891).csv\",low_memory=False)\n",
485
  "occurrence.info(show_counts=True)"
486
  ]
487
  },
@@ -1676,7 +1676,7 @@
1676
  "metadata": {},
1677
  "outputs": [],
1678
  "source": [
1679
- "df.to_csv(\"../deduplication_process/Jiggins_Zenodo_Img_Master_3477891Patch.csv\", index = False)"
1680
  ]
1681
  }
1682
  ],
 
34
  }
35
  ],
36
  "source": [
37
+ "multimedia = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/multimedia__(rec_3477891).csv\", low_memory=False)\n",
38
  "multimedia.info(show_counts=True)"
39
  ]
40
  },
 
481
  }
482
  ],
483
  "source": [
484
+ "occurrence = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/occurrences__(rec_3477891).csv\",low_memory=False)\n",
485
  "occurrence.info(show_counts=True)"
486
  ]
487
  },
 
1676
  "metadata": {},
1677
  "outputs": [],
1678
  "source": [
1679
+ "df.to_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv\", index = False)"
1680
  ]
1681
  }
1682
  ],
notebooks/Data-gen-1-2.ipynb CHANGED
@@ -15,8 +15,8 @@
15
  "source": [
16
  "## Summary\n",
17
  "\n",
18
- "We explored our downloaded images in `../deduplication_process/notebooks/EDA-DL-0-3.ipynb`, and established the following causes of duplication along with remedies:\n",
19
- "1. The \"patch\" record (record 3477891) which required alignment in `Data-gen-1-1.ipynb`, resulting in `../deduplication_process/Jiggins_Zenodo_Img_Master_3477891Patch.csv` used for download, is a duplication of all listed records. Additionally, it introduced extra copies of 4 images with incorrect labels for their views (i.e., dorsal vs ventral).\n",
20
  " - Solution: This record will be removed.\n",
21
  "2. The remaining duplication (430 images x2) comes from 5 records: [4291095](https://zenodo.org/records/4291095), [2813153](https://zenodo.org/records/2813153), [5526257](https://zenodo.org/records/5526257), [2553977](https://zenodo.org/records/2553977), and [2552371](https://zenodo.org/records/2552371).\n",
22
  " - Though there is nothing in their descriptions on Zenodo to indicate this overlap occurred, record 4291095 duplicated 415 images from record 2813153. 104 of these images had their view/side labels (dorsal vs ventral) updated to reflect an earlier mislabeling in record 2813153; these were all RAW copies of the images. All other metadata was consistent across records.\n",
@@ -37,7 +37,7 @@
37
  "\n",
38
  "## Generate \"Clean\" Master Metadata File\n",
39
  "\n",
40
- "We'll now generate a cleaned master metadata file from `../deduplication_process/metadata/Jiggins_Zenodo_Img_Master_3477891Patch_downloaded.csv` (generated in `../deduplication_process/notebooks/EDA-DL-0-3.ipynb` from our download master file and checksums of successful downloads) following the solutions outlined above. "
41
  ]
42
  },
43
  {
@@ -98,7 +98,7 @@
98
  }
99
  ],
100
  "source": [
101
- "df = pd.read_csv(\"../deduplication_process/metadata/Jiggins_Zenodo_Img_Master_3477891Patch_downloaded.csv\", low_memory = False)\n",
102
  "df[STATS_COLS].info()"
103
  ]
104
  },
 
15
  "source": [
16
  "## Summary\n",
17
  "\n",
18
+ "We explored our downloaded images in `EDA-DL-0-3.ipynb`, and established the following causes of duplication along with remedies:\n",
19
+ "1. The \"patch\" record (record 3477891) which required alignment in `Data-gen-1-1.ipynb`, resulting in `../metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv` used for download, is a duplication of all listed records. Additionally, it introduced extra copies of 4 images with incorrect labels for their views (i.e., dorsal vs ventral).\n",
20
  " - Solution: This record will be removed.\n",
21
  "2. The remaining duplication (430 images x2) comes from 5 records: [4291095](https://zenodo.org/records/4291095), [2813153](https://zenodo.org/records/2813153), [5526257](https://zenodo.org/records/5526257), [2553977](https://zenodo.org/records/2553977), and [2552371](https://zenodo.org/records/2552371).\n",
22
  " - Though there is nothing in their descriptions on Zenodo to indicate this overlap occurred, record 4291095 duplicated 415 images from record 2813153. 104 of these images had their view/side labels (dorsal vs ventral) updated to reflect an earlier mislabeling in record 2813153; these were all RAW copies of the images. All other metadata was consistent across records.\n",
 
37
  "\n",
38
  "## Generate \"Clean\" Master Metadata File\n",
39
  "\n",
40
+ "We'll now generate a cleaned master metadata file from `../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_downloaded.csv` (generated in `EDA-DL-0-3.ipynb` from our download master file and checksums of successful downloads) following the solutions outlined above. "
41
  ]
42
  },
43
  {
 
98
  }
99
  ],
100
  "source": [
101
+ "df = pd.read_csv(\"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_downloaded.csv\", low_memory = False)\n",
102
  "df[STATS_COLS].info()"
103
  ]
104
  },
notebooks/EDA-DL-0-1.ipynb CHANGED
@@ -27,7 +27,7 @@
27
  }
28
  ],
29
  "source": [
30
- "checksums = pd.read_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_checksums.csv\", low_memory=False)\n",
31
  "checksums.shape"
32
  ]
33
  },
@@ -77,7 +77,7 @@
77
  }
78
  ],
79
  "source": [
80
- "ERROR_PATH = \"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_error_log_(readable).json\"\n",
81
  "with open(ERROR_PATH, \"r\") as error_file:\n",
82
  " errors = json.loads(error_file.read())\n",
83
  "errors[:5]"
@@ -704,7 +704,7 @@
704
  "source": [
705
  "Is it possible some of these were recorded as part of this \"batch 2\" but are actually from batch 1 and that's why they're not there? \n",
706
  "\n",
707
- "Let's check the meta file `batch2.Peru.image.names.Zenodo.xlsx` (opened in Numbers and exported to CSV first, dropped `Zenodo` from name)."
708
  ]
709
  },
710
  {
@@ -800,7 +800,7 @@
800
  }
801
  ],
802
  "source": [
803
- "rec_4287444 = pd.read_csv(\"../metadata/Zenodo_meta_files/batch2.Peru.image.names.csv\", low_memory = True)\n",
804
  "rec_4287444.head()"
805
  ]
806
  },
@@ -1215,7 +1215,7 @@
1215
  "metadata": {},
1216
  "outputs": [],
1217
  "source": [
1218
- "master = pd.read_csv(\"../Jiggins_Zenodo_Img_Master_3477891Patch.csv\", low_memory = False)"
1219
  ]
1220
  },
1221
  {
@@ -2261,7 +2261,7 @@
2261
  }
2262
  ],
2263
  "source": [
2264
- "media_rec_3477891 = pd.read_csv(\"../metadata/Zenodo_meta_files/multimedia__(rec_3477891).csv\", low_memory=False)\n",
2265
  "\n",
2266
  "def get_link_filename(identifier):\n",
2267
  " if \"zenodo\" not in identifier:\n",
@@ -2793,7 +2793,7 @@
2793
  }
2794
  ],
2795
  "source": [
2796
- "LOG_PATH = \"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_log_(readable).json\"\n",
2797
  "\n",
2798
  "with open(LOG_PATH, \"r\") as log_file:\n",
2799
  " logs = json.loads(log_file.read())\n",
@@ -3057,7 +3057,7 @@
3057
  "metadata": {},
3058
  "outputs": [],
3059
  "source": [
3060
- "error_df.to_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_error_log.csv\", index = False)"
3061
  ]
3062
  },
3063
  {
 
27
  }
28
  ],
29
  "source": [
30
+ "checksums = pd.read_csv(\"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_checksums.csv\", low_memory=False)\n",
31
  "checksums.shape"
32
  ]
33
  },
 
77
  }
78
  ],
79
  "source": [
80
+ "ERROR_PATH = \"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_error_log_(readable).json\"\n",
81
  "with open(ERROR_PATH, \"r\") as error_file:\n",
82
  " errors = json.loads(error_file.read())\n",
83
  "errors[:5]"
 
704
  "source": [
705
  "Is it possible some of these were recorded as part of this \"batch 2\" but are actually from batch 1 and that's why they're not there? \n",
706
  "\n",
707
+ "Let's check the meta file: `batch2.Peru.image.names.Zenodo.xlsx`. This was opened in Numbers and exported to CSV first (using default `pd.read_excel` parameters resulted in an error trying to open the original Excel file), also dropped `Zenodo` from the filename."
708
  ]
709
  },
710
  {
 
800
  }
801
  ],
802
  "source": [
803
+ "rec_4287444 = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/batch2.Peru.image.names.csv\", low_memory = True)\n",
804
  "rec_4287444.head()"
805
  ]
806
  },
 
1215
  "metadata": {},
1216
  "outputs": [],
1217
  "source": [
1218
+ "master = pd.read_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv\", low_memory = False)"
1219
  ]
1220
  },
1221
  {
 
2261
  }
2262
  ],
2263
  "source": [
2264
+ "media_rec_3477891 = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/multimedia__(rec_3477891).csv\", low_memory=False)\n",
2265
  "\n",
2266
  "def get_link_filename(identifier):\n",
2267
  " if \"zenodo\" not in identifier:\n",
 
2793
  }
2794
  ],
2795
  "source": [
2796
+ "LOG_PATH = \"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_log_(readable).json\"\n",
2797
  "\n",
2798
  "with open(LOG_PATH, \"r\") as log_file:\n",
2799
  " logs = json.loads(log_file.read())\n",
 
3057
  "metadata": {},
3058
  "outputs": [],
3059
  "source": [
3060
+ "error_df.to_csv(\"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_error_log.csv\", index = False)"
3061
  ]
3062
  },
3063
  {
notebooks/EDA-DL-0-2.ipynb CHANGED
@@ -36,7 +36,7 @@
36
  "source": [
37
  "# Record 4289223 is most of our initial 404s (ultimately just second most), \n",
38
  "# looks like they align with 0.2.rachel.blow.mel.tim.ikiam.csv, so we'll check\n",
39
- "fof_df = pd.read_csv(\"../metadata/Zenodo_meta_files/0.2.rachel.blow.mel.tim.ikiam.csv\", low_memory = False)\n",
40
  "fof_df.info(show_counts=True)"
41
  ]
42
  },
@@ -250,7 +250,7 @@
250
  }
251
  ],
252
  "source": [
253
- "ERROR_PATH = \"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_error_log_(readable).json\"\n",
254
  "with open(ERROR_PATH, \"r\") as error_file:\n",
255
  " errors = json.loads(error_file.read())\n",
256
  "\n",
@@ -347,10 +347,10 @@
347
  }
348
  ],
349
  "source": [
350
- "error_df = pd.read_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_error_log.csv\", low_memory = False)\n",
351
  "print(error_df.columns)\n",
352
  "\n",
353
- "master = pd.read_csv(\"../Jiggins_Zenodo_Img_Master_3477891Patch.csv\", low_memory = False)"
354
  ]
355
  },
356
  {
 
36
  "source": [
37
  "# Record 4289223 is most of our initial 404s (ultimately just second most), \n",
38
  "# looks like they align with 0.2.rachel.blow.mel.tim.ikiam.csv, so we'll check\n",
39
+ "fof_df = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/0.2.rachel.blow.mel.tim.ikiam.csv\", low_memory = False)\n",
40
  "fof_df.info(show_counts=True)"
41
  ]
42
  },
 
250
  }
251
  ],
252
  "source": [
253
+ "ERROR_PATH = \"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_error_log_(readable).json\"\n",
254
  "with open(ERROR_PATH, \"r\") as error_file:\n",
255
  " errors = json.loads(error_file.read())\n",
256
  "\n",
 
347
  }
348
  ],
349
  "source": [
350
+ "error_df = pd.read_csv(\"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_error_log.csv\", low_memory = False)\n",
351
  "print(error_df.columns)\n",
352
  "\n",
353
+ "master = pd.read_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv\", low_memory = False)"
354
  ]
355
  },
356
  {
notebooks/EDA-DL-0-3.ipynb CHANGED
@@ -38,7 +38,7 @@
38
  }
39
  ],
40
  "source": [
41
- "checksums = pd.read_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_checksums.csv\", low_memory=False)\n",
42
  "checksums.shape"
43
  ]
44
  },
@@ -89,7 +89,7 @@
89
  }
90
  ],
91
  "source": [
92
- "master = pd.read_csv(\"../Jiggins_Zenodo_Img_Master_3477891Patch.csv\", low_memory = False)\n",
93
  "master.shape"
94
  ]
95
  },
@@ -489,7 +489,7 @@
489
  "outputs": [],
490
  "source": [
491
  "# Save master file of all downloaded images with MD5s (no duplicate marker)\n",
492
- "master_checksums[list(master_checksums.columns)[:-1]].to_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch_downloaded.csv\", index = False)"
493
  ]
494
  },
495
  {
@@ -1939,8 +1939,8 @@
1939
  "metadata": {},
1940
  "outputs": [],
1941
  "source": [
1942
- "df_4291095 = pd.read_csv(\"../metadata/Zenodo_meta_files/0.gmk.broods.all_(rec4291095).csv\", low_memory=False)\n",
1943
- "df_2813153 = pd.read_csv(\"../metadata/Zenodo_meta_files/CAM.coll.images.batch10_(rec2813153).csv\", low_memory=False)"
1944
  ]
1945
  },
1946
  {
 
38
  }
39
  ],
40
  "source": [
41
+ "checksums = pd.read_csv(\"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_checksums.csv\", low_memory=False)\n",
42
  "checksums.shape"
43
  ]
44
  },
 
89
  }
90
  ],
91
  "source": [
92
+ "master = pd.read_csv(\"../metadata/Jiggins_Zenodo_Img_Master_3477891Patch.csv\", low_memory = False)\n",
93
  "master.shape"
94
  ]
95
  },
 
489
  "outputs": [],
490
  "source": [
491
  "# Save master file of all downloaded images with MD5s (no duplicate marker)\n",
492
+ "master_checksums[list(master_checksums.columns)[:-1]].to_csv(\"../metadata/deduplication/Jiggins_Zenodo_Img_Master_3477891Patch_downloaded.csv\", index = False)"
493
  ]
494
  },
495
  {
 
1939
  "metadata": {},
1940
  "outputs": [],
1941
  "source": [
1942
+ "df_4291095 = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/0.gmk.broods.all_(rec4291095).csv\", low_memory=False)\n",
1943
+ "df_2813153 = pd.read_csv(\"../metadata/deduplication/Zenodo_meta_files/CAM.coll.images.batch10_(rec2813153).csv\", low_memory=False)"
1944
  ]
1945
  },
1946
  {