Skip to content

Commit d761e59

Browse files
authored
Merge pull request #167 from dnth/dnth/update-example
Rerun broken notebooks
2 parents 9697e66 + b4da1cc commit d761e59

File tree

2 files changed

+1937
-1329
lines changed

2 files changed

+1937
-1329
lines changed

examples/analysing-image-classification-dataset.ipynb

Lines changed: 55 additions & 48 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,6 @@
11
{
22
"cells": [
33
{
4-
"attachments": {},
54
"cell_type": "markdown",
65
"id": "SwSYWR4vzk_e",
76
"metadata": {
@@ -149,7 +148,6 @@
149148
]
150149
},
151150
{
152-
"attachments": {},
153151
"cell_type": "markdown",
154152
"id": "8aba34e1",
155153
"metadata": {},
@@ -247,8 +245,8 @@
247245
"</div>"
248246
],
249247
"text/plain": [
250-
" path noisy_labels_0 noisy_labels_1 \\\n",
251-
"0 train/n02979186/n02979186_9036.JPEG n02979186 n02979186 \n",
248+
" path noisy_labels_0 noisy_labels_1 \n",
249+
"0 train/n02979186/n02979186_9036.JPEG n02979186 n02979186 \\\n",
252250
"1 train/n02979186/n02979186_11957.JPEG n02979186 n02979186 \n",
253251
"2 train/n02979186/n02979186_9715.JPEG n02979186 n02979186 \n",
254252
"\n",
@@ -269,7 +267,6 @@
269267
]
270268
},
271269
{
272-
"attachments": {},
273270
"cell_type": "markdown",
274271
"id": "dfc957bf",
275272
"metadata": {},
@@ -322,41 +319,41 @@
322319
" <thead>\n",
323320
" <tr style=\"text-align: right;\">\n",
324321
" <th></th>\n",
325-
" <th>img_filename</th>\n",
322+
" <th>filename</th>\n",
326323
" <th>label</th>\n",
327324
" <th>split</th>\n",
328325
" </tr>\n",
329326
" </thead>\n",
330327
" <tbody>\n",
331328
" <tr>\n",
332329
" <th>0</th>\n",
333-
" <td>train/n02979186/n02979186_9036.JPEG</td>\n",
330+
" <td>imagenette2-160/train/n02979186/n02979186_9036...</td>\n",
334331
" <td>cassette_player</td>\n",
335-
" <td>train</td>\n",
332+
" <td>imagenette2-160</td>\n",
336333
" </tr>\n",
337334
" <tr>\n",
338335
" <th>1</th>\n",
339-
" <td>train/n02979186/n02979186_11957.JPEG</td>\n",
336+
" <td>imagenette2-160/train/n02979186/n02979186_1195...</td>\n",
340337
" <td>cassette_player</td>\n",
341-
" <td>train</td>\n",
338+
" <td>imagenette2-160</td>\n",
342339
" </tr>\n",
343340
" <tr>\n",
344341
" <th>2</th>\n",
345-
" <td>train/n02979186/n02979186_9715.JPEG</td>\n",
342+
" <td>imagenette2-160/train/n02979186/n02979186_9715...</td>\n",
346343
" <td>cassette_player</td>\n",
347-
" <td>train</td>\n",
344+
" <td>imagenette2-160</td>\n",
348345
" </tr>\n",
349346
" <tr>\n",
350347
" <th>3</th>\n",
351-
" <td>train/n02979186/n02979186_21736.JPEG</td>\n",
348+
" <td>imagenette2-160/train/n02979186/n02979186_2173...</td>\n",
352349
" <td>cassette_player</td>\n",
353-
" <td>train</td>\n",
350+
" <td>imagenette2-160</td>\n",
354351
" </tr>\n",
355352
" <tr>\n",
356353
" <th>4</th>\n",
357-
" <td>train/n02979186/ILSVRC2012_val_00046953.JPEG</td>\n",
354+
" <td>imagenette2-160/train/n02979186/ILSVRC2012_val...</td>\n",
358355
" <td>cassette_player</td>\n",
359-
" <td>train</td>\n",
356+
" <td>imagenette2-160</td>\n",
360357
" </tr>\n",
361358
" <tr>\n",
362359
" <th>...</th>\n",
@@ -366,52 +363,65 @@
366363
" </tr>\n",
367364
" <tr>\n",
368365
" <th>13389</th>\n",
369-
" <td>val/n03425413/n03425413_17521.JPEG</td>\n",
366+
" <td>imagenette2-160/val/n03425413/n03425413_17521....</td>\n",
370367
" <td>gas_pump</td>\n",
371-
" <td>val</td>\n",
368+
" <td>imagenette2-160</td>\n",
372369
" </tr>\n",
373370
" <tr>\n",
374371
" <th>13390</th>\n",
375-
" <td>val/n03425413/n03425413_20711.JPEG</td>\n",
372+
" <td>imagenette2-160/val/n03425413/n03425413_20711....</td>\n",
376373
" <td>gas_pump</td>\n",
377-
" <td>val</td>\n",
374+
" <td>imagenette2-160</td>\n",
378375
" </tr>\n",
379376
" <tr>\n",
380377
" <th>13391</th>\n",
381-
" <td>val/n03425413/n03425413_19050.JPEG</td>\n",
378+
" <td>imagenette2-160/val/n03425413/n03425413_19050....</td>\n",
382379
" <td>gas_pump</td>\n",
383-
" <td>val</td>\n",
380+
" <td>imagenette2-160</td>\n",
384381
" </tr>\n",
385382
" <tr>\n",
386383
" <th>13392</th>\n",
387-
" <td>val/n03425413/n03425413_13831.JPEG</td>\n",
384+
" <td>imagenette2-160/val/n03425413/n03425413_13831....</td>\n",
388385
" <td>gas_pump</td>\n",
389-
" <td>val</td>\n",
386+
" <td>imagenette2-160</td>\n",
390387
" </tr>\n",
391388
" <tr>\n",
392389
" <th>13393</th>\n",
393-
" <td>val/n03425413/n03425413_1242.JPEG</td>\n",
390+
" <td>imagenette2-160/val/n03425413/n03425413_1242.JPEG</td>\n",
394391
" <td>gas_pump</td>\n",
395-
" <td>val</td>\n",
392+
" <td>imagenette2-160</td>\n",
396393
" </tr>\n",
397394
" </tbody>\n",
398395
"</table>\n",
399396
"<p>13394 rows × 3 columns</p>\n",
400397
"</div>"
401398
],
402399
"text/plain": [
403-
" img_filename label split\n",
404-
"0 train/n02979186/n02979186_9036.JPEG cassette_player train\n",
405-
"1 train/n02979186/n02979186_11957.JPEG cassette_player train\n",
406-
"2 train/n02979186/n02979186_9715.JPEG cassette_player train\n",
407-
"3 train/n02979186/n02979186_21736.JPEG cassette_player train\n",
408-
"4 train/n02979186/ILSVRC2012_val_00046953.JPEG cassette_player train\n",
409-
"... ... ... ...\n",
410-
"13389 val/n03425413/n03425413_17521.JPEG gas_pump val\n",
411-
"13390 val/n03425413/n03425413_20711.JPEG gas_pump val\n",
412-
"13391 val/n03425413/n03425413_19050.JPEG gas_pump val\n",
413-
"13392 val/n03425413/n03425413_13831.JPEG gas_pump val\n",
414-
"13393 val/n03425413/n03425413_1242.JPEG gas_pump val\n",
400+
" filename label \n",
401+
"0 imagenette2-160/train/n02979186/n02979186_9036... cassette_player \\\n",
402+
"1 imagenette2-160/train/n02979186/n02979186_1195... cassette_player \n",
403+
"2 imagenette2-160/train/n02979186/n02979186_9715... cassette_player \n",
404+
"3 imagenette2-160/train/n02979186/n02979186_2173... cassette_player \n",
405+
"4 imagenette2-160/train/n02979186/ILSVRC2012_val... cassette_player \n",
406+
"... ... ... \n",
407+
"13389 imagenette2-160/val/n03425413/n03425413_17521.... gas_pump \n",
408+
"13390 imagenette2-160/val/n03425413/n03425413_20711.... gas_pump \n",
409+
"13391 imagenette2-160/val/n03425413/n03425413_19050.... gas_pump \n",
410+
"13392 imagenette2-160/val/n03425413/n03425413_13831.... gas_pump \n",
411+
"13393 imagenette2-160/val/n03425413/n03425413_1242.JPEG gas_pump \n",
412+
"\n",
413+
" split \n",
414+
"0 imagenette2-160 \n",
415+
"1 imagenette2-160 \n",
416+
"2 imagenette2-160 \n",
417+
"3 imagenette2-160 \n",
418+
"4 imagenette2-160 \n",
419+
"... ... \n",
420+
"13389 imagenette2-160 \n",
421+
"13390 imagenette2-160 \n",
422+
"13391 imagenette2-160 \n",
423+
"13392 imagenette2-160 \n",
424+
"13393 imagenette2-160 \n",
415425
"\n",
416426
"[13394 rows x 3 columns]"
417427
]
@@ -428,6 +438,9 @@
428438
"# rename columns to fastdup's column names\n",
429439
"df_annot = df_annot.rename({'noisy_labels_0': 'label', 'path': 'filename'}, axis='columns')\n",
430440
"\n",
441+
"# append datadir\n",
442+
"df_annot['filename'] = df_annot['filename'].apply(lambda x: data_dir + x)\n",
443+
"\n",
431444
"# create split column\n",
432445
"df_annot['split'] = df_annot['filename'].apply(lambda x: x.split(\"/\")[0])\n",
433446
"\n",
@@ -439,7 +452,6 @@
439452
]
440453
},
441454
{
442-
"attachments": {},
443455
"cell_type": "markdown",
444456
"id": "0c648ed1-5016-4230-9873-546eb510b764",
445457
"metadata": {
@@ -453,17 +465,17 @@
453465
},
454466
{
455467
"cell_type": "code",
456-
"execution_count": 1,
468+
"execution_count": 6,
457469
"id": "7f69d8b2",
458470
"metadata": {},
459471
"outputs": [
460472
{
461473
"data": {
462474
"text/plain": [
463-
"'0.918'"
475+
"'0.922'"
464476
]
465477
},
466-
"execution_count": 1,
478+
"execution_count": 6,
467479
"metadata": {},
468480
"output_type": "execute_result"
469481
}
@@ -522,7 +534,6 @@
522534
]
523535
},
524536
{
525-
"attachments": {},
526537
"cell_type": "markdown",
527538
"id": "62e35a12-fadd-4b3f-bcab-69e6e67862a4",
528539
"metadata": {},
@@ -1662,7 +1673,6 @@
16621673
]
16631674
},
16641675
{
1665-
"attachments": {},
16661676
"cell_type": "markdown",
16671677
"id": "67378b58",
16681678
"metadata": {},
@@ -3998,7 +4008,6 @@
39984008
]
39994009
},
40004010
{
4001-
"attachments": {},
40024011
"cell_type": "markdown",
40034012
"id": "c2c393be-2b42-4814-8688-03d2be9e8998",
40044013
"metadata": {},
@@ -4833,7 +4842,6 @@
48334842
]
48344843
},
48354844
{
4836-
"attachments": {},
48374845
"cell_type": "markdown",
48384846
"id": "e10989e1",
48394847
"metadata": {},
@@ -7707,7 +7715,6 @@
77077715
]
77087716
},
77097717
{
7710-
"attachments": {},
77117718
"cell_type": "markdown",
77127719
"id": "569cb878",
77137720
"metadata": {},
@@ -7778,7 +7785,7 @@
77787785
"name": "python",
77797786
"nbconvert_exporter": "python",
77807787
"pygments_lexer": "ipython3",
7781-
"version": "3.10.9"
7788+
"version": "3.10.11"
77827789
}
77837790
},
77847791
"nbformat": 4,

0 commit comments

Comments
 (0)