Hugging Face's logo

mnjm
/
deit-b16-food101

Image Classification
PyTorch
food101
vision-transformer
deit
Model card Files
xet
 Community
mnjm commited on
Commit
844452c
·
verified ·
1 Parent(s): 7443339

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. DeiT-B-16-food101.log +242 -0
  2. DeiT-B-16-food101.pt +3 -0
  3. README.md +40 -0
  4. config/config.yaml +63 -0
  5. config/hydra.yaml +159 -0
  6. config/overrides.yaml +2 -0
DeiT-B-16-food101.log ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2026-07-09 23:27:41,431][vit][INFO] - Using cuda
2
+ [2026-07-09 23:27:41,432][vit][INFO] - Loading food101 dataset
3
+ [2026-07-09 23:27:42,003][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
4
+ [2026-07-09 23:27:42,012][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/ethz/food101/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/README.md "HTTP/1.1 200 OK"
5
+ [2026-07-09 23:27:42,246][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/food101.py "HTTP/1.1 404 Not Found"
6
+ [2026-07-09 23:27:44,530][httpx][INFO] - HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/ethz/food101/ethz/food101.py "HTTP/1.1 404 Not Found"
7
+ [2026-07-09 23:27:51,463][httpx][INFO] - HTTP Request: GET https://huggingface.co/api/datasets/ethz/food101/revision/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9 "HTTP/1.1 200 OK"
8
+ [2026-07-09 23:27:51,701][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/.huggingface.yaml "HTTP/1.1 404 Not Found"
9
+ [2026-07-09 23:27:51,985][httpx][INFO] - HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=ethz/food101 "HTTP/1.1 200 OK"
10
+ [2026-07-09 23:27:52,412][httpx][INFO] - HTTP Request: GET https://huggingface.co/api/datasets/ethz/food101/tree/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/data?recursive=true&expand=false "HTTP/1.1 200 OK"
11
+ [2026-07-09 23:28:01,665][httpx][INFO] - HTTP Request: GET https://huggingface.co/api/datasets/ethz/food101/tree/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9?recursive=false&expand=false "HTTP/1.1 200 OK"
12
+ [2026-07-09 23:28:02,166][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/dataset_infos.json "HTTP/1.1 404 Not Found"
13
+ [2026-07-09 23:28:02,424][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
14
+ [2026-07-09 23:28:02,432][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/ethz/food101/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/README.md "HTTP/1.1 200 OK"
15
+ [2026-07-09 23:28:03,052][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/food101.py "HTTP/1.1 404 Not Found"
16
+ [2026-07-09 23:28:03,798][httpx][INFO] - HTTP Request: HEAD https://s3.amazonaws.com/datasets.huggingface.co/datasets/datasets/ethz/food101/ethz/food101.py "HTTP/1.1 404 Not Found"
17
+ [2026-07-09 23:28:04,046][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/.huggingface.yaml "HTTP/1.1 404 Not Found"
18
+ [2026-07-09 23:28:04,304][httpx][INFO] - HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=ethz/food101 "HTTP/1.1 200 OK"
19
+ [2026-07-09 23:28:04,540][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/dataset_infos.json "HTTP/1.1 404 Not Found"
20
+ [2026-07-09 23:28:06,831][vit][INFO] - Model type: DeiT-B-16 params: 85,955,530
21
+ [2026-07-09 23:28:10,829][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/anonauthors/food101-resnet50/resolve/main/config.json "HTTP/1.1 307 Temporary Redirect"
22
+ [2026-07-09 23:28:11,480][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/api/resolve-cache/models/anonauthors/food101-resnet50/0e0c5e7df53238af8435d256bce42dd4ef7be5c6/config.json "HTTP/1.1 200 OK"
23
+ [2026-07-09 23:28:11,560][timm.models._builder][INFO] - Loading pretrained weights from Hugging Face hub (anonauthors/food101-resnet50)
24
+ [2026-07-09 23:28:11,914][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/anonauthors/food101-resnet50/resolve/main/model.safetensors "HTTP/1.1 404 Not Found"
25
+ [2026-07-09 23:28:12,390][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/anonauthors/food101-resnet50/resolve/main/pytorch_model.bin "HTTP/1.1 302 Found"
26
+ [2026-07-09 23:28:12,877][vit][INFO] - Loaded teacher model teacher_name='hf_hub:anonauthors/food101-resnet50'
27
+ [2026-07-09 23:28:56,476][vit][INFO] - Initial Val loss=4.7742 acc@1=0.0087 acc@5=0.0450 Time=35.34s
28
+ [2026-07-09 23:28:56,477][vit][INFO] - Epoch: 1/100
29
+ [2026-07-09 23:30:32,708][vit][INFO] - Train loss=4.4966 acc@1=0.0210 acc@5=0.0852 Time=96.23s
30
+ [2026-07-09 23:30:32,709][vit][INFO] - Epoch: 2/100
31
+ [2026-07-09 23:31:48,710][vit][INFO] - Train loss=4.1817 acc@1=0.0568 acc@5=0.1827 Time=76.00s
32
+ [2026-07-09 23:31:48,711][vit][INFO] - Epoch: 3/100
33
+ [2026-07-09 23:33:05,109][vit][INFO] - Train loss=3.8907 acc@1=0.1003 acc@5=0.2776 Time=76.40s
34
+ [2026-07-09 23:33:05,110][vit][INFO] - Epoch: 4/100
35
+ [2026-07-09 23:34:21,896][vit][INFO] - Train loss=3.6744 acc@1=0.1335 acc@5=0.3427 Time=76.79s
36
+ [2026-07-09 23:34:21,897][vit][INFO] - Epoch: 5/100
37
+ [2026-07-09 23:35:38,931][vit][INFO] - Train loss=3.5467 acc@1=0.1580 acc@5=0.3816 Time=77.03s
38
+ [2026-07-09 23:35:38,932][vit][INFO] - Epoch: 6/100
39
+ [2026-07-09 23:36:56,059][vit][INFO] - Train loss=3.4199 acc@1=0.1843 acc@5=0.4201 Time=77.13s
40
+ [2026-07-09 23:36:56,060][vit][INFO] - Epoch: 7/100
41
+ [2026-07-09 23:38:13,034][vit][INFO] - Train loss=3.3239 acc@1=0.2021 acc@5=0.4450 Time=76.97s
42
+ [2026-07-09 23:38:13,035][vit][INFO] - Epoch: 8/100
43
+ [2026-07-09 23:39:29,907][vit][INFO] - Train loss=3.2359 acc@1=0.2196 acc@5=0.4700 Time=76.87s
44
+ [2026-07-09 23:39:29,908][vit][INFO] - Epoch: 9/100
45
+ [2026-07-09 23:40:46,902][vit][INFO] - Train loss=3.1513 acc@1=0.2374 acc@5=0.4933 Time=76.99s
46
+ [2026-07-09 23:40:46,902][vit][INFO] - Epoch: 10/100
47
+ [2026-07-09 23:42:03,916][vit][INFO] - Train loss=3.0945 acc@1=0.2502 acc@5=0.5101 Time=77.01s
48
+ [2026-07-09 23:42:18,416][vit][INFO] - Val loss=2.7778 acc@1=0.3230 acc@5=0.6131 Time=14.50s
49
+ [2026-07-09 23:42:18,417][vit][INFO] - Epoch: 11/100
50
+ [2026-07-09 23:43:35,371][vit][INFO] - Train loss=3.0081 acc@1=0.2669 acc@5=0.5329 Time=76.95s
51
+ [2026-07-09 23:43:35,372][vit][INFO] - Epoch: 12/100
52
+ [2026-07-09 23:44:52,312][vit][INFO] - Train loss=2.9252 acc@1=0.2849 acc@5=0.5511 Time=76.94s
53
+ [2026-07-09 23:44:52,313][vit][INFO] - Epoch: 13/100
54
+ [2026-07-09 23:46:09,687][vit][INFO] - Train loss=2.8202 acc@1=0.3085 acc@5=0.5790 Time=77.37s
55
+ [2026-07-09 23:46:09,687][vit][INFO] - Epoch: 14/100
56
+ [2026-07-09 23:47:26,644][vit][INFO] - Train loss=2.7328 acc@1=0.3279 acc@5=0.5992 Time=76.96s
57
+ [2026-07-09 23:47:26,645][vit][INFO] - Epoch: 15/100
58
+ [2026-07-09 23:48:43,513][vit][INFO] - Train loss=2.6584 acc@1=0.3434 acc@5=0.6195 Time=76.87s
59
+ [2026-07-09 23:48:43,514][vit][INFO] - Epoch: 16/100
60
+ [2026-07-09 23:50:00,650][vit][INFO] - Train loss=2.5653 acc@1=0.3674 acc@5=0.6407 Time=77.14s
61
+ [2026-07-09 23:50:00,650][vit][INFO] - Epoch: 17/100
62
+ [2026-07-09 23:51:17,621][vit][INFO] - Train loss=2.4803 acc@1=0.3878 acc@5=0.6614 Time=76.97s
63
+ [2026-07-09 23:51:17,622][vit][INFO] - Epoch: 18/100
64
+ [2026-07-09 23:52:34,671][vit][INFO] - Train loss=2.4044 acc@1=0.4058 acc@5=0.6763 Time=77.05s
65
+ [2026-07-09 23:52:34,672][vit][INFO] - Epoch: 19/100
66
+ [2026-07-09 23:53:51,569][vit][INFO] - Train loss=2.3306 acc@1=0.4240 acc@5=0.6942 Time=76.90s
67
+ [2026-07-09 23:53:51,570][vit][INFO] - Epoch: 20/100
68
+ [2026-07-09 23:55:08,403][vit][INFO] - Train loss=2.2378 acc@1=0.4468 acc@5=0.7134 Time=76.83s
69
+ [2026-07-09 23:55:22,842][vit][INFO] - Val loss=2.0015 acc@1=0.5080 acc@5=0.7784 Time=14.44s
70
+ [2026-07-09 23:55:23,287][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/DeiT-B-16-food101-09-07-26-232741/DeiT-B-16-food101.pt
71
+ [2026-07-09 23:55:23,287][vit][INFO] - Epoch: 21/100
72
+ [2026-07-09 23:56:40,084][vit][INFO] - Train loss=2.1600 acc@1=0.4644 acc@5=0.7309 Time=76.80s
73
+ [2026-07-09 23:56:40,084][vit][INFO] - Epoch: 22/100
74
+ [2026-07-09 23:57:56,939][vit][INFO] - Train loss=2.0836 acc@1=0.4868 acc@5=0.7494 Time=76.85s
75
+ [2026-07-09 23:57:56,940][vit][INFO] - Epoch: 23/100
76
+ [2026-07-09 23:59:13,881][vit][INFO] - Train loss=2.0037 acc@1=0.5057 acc@5=0.7650 Time=76.94s
77
+ [2026-07-09 23:59:13,882][vit][INFO] - Epoch: 24/100
78
+ [2026-07-10 00:00:31,054][vit][INFO] - Train loss=1.9131 acc@1=0.5272 acc@5=0.7843 Time=77.17s
79
+ [2026-07-10 00:00:31,054][vit][INFO] - Epoch: 25/100
80
+ [2026-07-10 00:01:47,980][vit][INFO] - Train loss=1.8362 acc@1=0.5475 acc@5=0.7996 Time=76.93s
81
+ [2026-07-10 00:01:47,981][vit][INFO] - Epoch: 26/100
82
+ [2026-07-10 00:03:04,795][vit][INFO] - Train loss=1.7597 acc@1=0.5727 acc@5=0.8163 Time=76.81s
83
+ [2026-07-10 00:03:04,796][vit][INFO] - Epoch: 27/100
84
+ [2026-07-10 00:04:21,642][vit][INFO] - Train loss=1.6713 acc@1=0.5945 acc@5=0.8324 Time=76.85s
85
+ [2026-07-10 00:04:21,642][vit][INFO] - Epoch: 28/100
86
+ [2026-07-10 00:05:38,582][vit][INFO] - Train loss=1.5940 acc@1=0.6171 acc@5=0.8478 Time=76.94s
87
+ [2026-07-10 00:05:38,583][vit][INFO] - Epoch: 29/100
88
+ [2026-07-10 00:06:55,479][vit][INFO] - Train loss=1.5161 acc@1=0.6420 acc@5=0.8624 Time=76.90s
89
+ [2026-07-10 00:06:55,480][vit][INFO] - Epoch: 30/100
90
+ [2026-07-10 00:08:12,184][vit][INFO] - Train loss=1.4474 acc@1=0.6630 acc@5=0.8774 Time=76.70s
91
+ [2026-07-10 00:08:26,465][vit][INFO] - Val loss=1.8461 acc@1=0.5757 acc@5=0.8220 Time=14.28s
92
+ [2026-07-10 00:08:26,466][vit][INFO] - Epoch: 31/100
93
+ [2026-07-10 00:09:43,119][vit][INFO] - Train loss=1.3725 acc@1=0.6870 acc@5=0.8873 Time=76.65s
94
+ [2026-07-10 00:09:43,120][vit][INFO] - Epoch: 32/100
95
+ [2026-07-10 00:10:59,942][vit][INFO] - Train loss=1.3142 acc@1=0.7066 acc@5=0.8971 Time=76.82s
96
+ [2026-07-10 00:10:59,943][vit][INFO] - Epoch: 33/100
97
+ [2026-07-10 00:12:16,988][vit][INFO] - Train loss=1.2499 acc@1=0.7252 acc@5=0.9105 Time=77.05s
98
+ [2026-07-10 00:12:16,989][vit][INFO] - Epoch: 34/100
99
+ [2026-07-10 00:13:33,811][vit][INFO] - Train loss=1.1944 acc@1=0.7431 acc@5=0.9176 Time=76.82s
100
+ [2026-07-10 00:13:33,811][vit][INFO] - Epoch: 35/100
101
+ [2026-07-10 00:14:50,928][vit][INFO] - Train loss=1.1468 acc@1=0.7618 acc@5=0.9242 Time=77.12s
102
+ [2026-07-10 00:14:50,929][vit][INFO] - Epoch: 36/100
103
+ [2026-07-10 00:16:07,752][vit][INFO] - Train loss=1.0874 acc@1=0.7798 acc@5=0.9332 Time=76.82s
104
+ [2026-07-10 00:16:07,753][vit][INFO] - Epoch: 37/100
105
+ [2026-07-10 00:17:24,547][vit][INFO] - Train loss=1.0481 acc@1=0.7918 acc@5=0.9382 Time=76.79s
106
+ [2026-07-10 00:17:24,547][vit][INFO] - Epoch: 38/100
107
+ [2026-07-10 00:18:41,278][vit][INFO] - Train loss=1.0088 acc@1=0.8067 acc@5=0.9440 Time=76.73s
108
+ [2026-07-10 00:18:41,279][vit][INFO] - Epoch: 39/100
109
+ [2026-07-10 00:19:58,092][vit][INFO] - Train loss=0.9782 acc@1=0.8173 acc@5=0.9480 Time=76.81s
110
+ [2026-07-10 00:19:58,093][vit][INFO] - Epoch: 40/100
111
+ [2026-07-10 00:21:14,986][vit][INFO] - Train loss=0.9422 acc@1=0.8292 acc@5=0.9536 Time=76.89s
112
+ [2026-07-10 00:21:29,059][vit][INFO] - Val loss=1.9594 acc@1=0.5905 acc@5=0.8283 Time=14.07s
113
+ [2026-07-10 00:21:29,708][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/DeiT-B-16-food101-09-07-26-232741/DeiT-B-16-food101.pt
114
+ [2026-07-10 00:21:29,709][vit][INFO] - Epoch: 41/100
115
+ [2026-07-10 00:22:46,483][vit][INFO] - Train loss=0.9189 acc@1=0.8382 acc@5=0.9539 Time=76.77s
116
+ [2026-07-10 00:22:46,483][vit][INFO] - Epoch: 42/100
117
+ [2026-07-10 00:24:03,248][vit][INFO] - Train loss=0.8832 acc@1=0.8508 acc@5=0.9600 Time=76.76s
118
+ [2026-07-10 00:24:03,249][vit][INFO] - Epoch: 43/100
119
+ [2026-07-10 00:25:20,040][vit][INFO] - Train loss=0.8561 acc@1=0.8604 acc@5=0.9634 Time=76.79s
120
+ [2026-07-10 00:25:20,041][vit][INFO] - Epoch: 44/100
121
+ [2026-07-10 00:26:36,900][vit][INFO] - Train loss=0.8299 acc@1=0.8665 acc@5=0.9655 Time=76.86s
122
+ [2026-07-10 00:26:36,901][vit][INFO] - Epoch: 45/100
123
+ [2026-07-10 00:27:53,791][vit][INFO] - Train loss=0.8097 acc@1=0.8738 acc@5=0.9670 Time=76.89s
124
+ [2026-07-10 00:27:53,791][vit][INFO] - Epoch: 46/100
125
+ [2026-07-10 00:29:10,727][vit][INFO] - Train loss=0.7836 acc@1=0.8825 acc@5=0.9709 Time=76.94s
126
+ [2026-07-10 00:29:10,728][vit][INFO] - Epoch: 47/100
127
+ [2026-07-10 00:30:28,213][vit][INFO] - Train loss=0.7688 acc@1=0.8876 acc@5=0.9720 Time=77.49s
128
+ [2026-07-10 00:30:28,214][vit][INFO] - Epoch: 48/100
129
+ [2026-07-10 00:31:44,884][vit][INFO] - Train loss=0.7485 acc@1=0.8953 acc@5=0.9747 Time=76.67s
130
+ [2026-07-10 00:31:44,885][vit][INFO] - Epoch: 49/100
131
+ [2026-07-10 00:33:01,771][vit][INFO] - Train loss=0.7341 acc@1=0.9007 acc@5=0.9752 Time=76.89s
132
+ [2026-07-10 00:33:01,772][vit][INFO] - Epoch: 50/100
133
+ [2026-07-10 00:34:18,747][vit][INFO] - Train loss=0.7215 acc@1=0.9044 acc@5=0.9762 Time=76.97s
134
+ [2026-07-10 00:34:33,155][vit][INFO] - Val loss=1.9602 acc@1=0.6029 acc@5=0.8343 Time=14.41s
135
+ [2026-07-10 00:34:33,155][vit][INFO] - Epoch: 51/100
136
+ [2026-07-10 00:35:49,924][vit][INFO] - Train loss=0.6969 acc@1=0.9093 acc@5=0.9779 Time=76.77s
137
+ [2026-07-10 00:35:49,925][vit][INFO] - Epoch: 52/100
138
+ [2026-07-10 00:37:06,625][vit][INFO] - Train loss=0.6911 acc@1=0.9131 acc@5=0.9798 Time=76.70s
139
+ [2026-07-10 00:37:06,626][vit][INFO] - Epoch: 53/100
140
+ [2026-07-10 00:38:23,378][vit][INFO] - Train loss=0.6731 acc@1=0.9192 acc@5=0.9820 Time=76.75s
141
+ [2026-07-10 00:38:23,378][vit][INFO] - Epoch: 54/100
142
+ [2026-07-10 00:39:40,175][vit][INFO] - Train loss=0.6585 acc@1=0.9224 acc@5=0.9823 Time=76.80s
143
+ [2026-07-10 00:39:40,176][vit][INFO] - Epoch: 55/100
144
+ [2026-07-10 00:40:57,045][vit][INFO] - Train loss=0.6451 acc@1=0.9256 acc@5=0.9837 Time=76.87s
145
+ [2026-07-10 00:40:57,045][vit][INFO] - Epoch: 56/100
146
+ [2026-07-10 00:42:13,837][vit][INFO] - Train loss=0.6341 acc@1=0.9278 acc@5=0.9844 Time=76.79s
147
+ [2026-07-10 00:42:13,838][vit][INFO] - Epoch: 57/100
148
+ [2026-07-10 00:43:30,712][vit][INFO] - Train loss=0.6296 acc@1=0.9318 acc@5=0.9856 Time=76.87s
149
+ [2026-07-10 00:43:30,712][vit][INFO] - Epoch: 58/100
150
+ [2026-07-10 00:44:47,733][vit][INFO] - Train loss=0.6190 acc@1=0.9347 acc@5=0.9863 Time=77.02s
151
+ [2026-07-10 00:44:47,735][vit][INFO] - Epoch: 59/100
152
+ [2026-07-10 00:46:04,622][vit][INFO] - Train loss=0.6065 acc@1=0.9377 acc@5=0.9865 Time=76.84s
153
+ [2026-07-10 00:46:04,623][vit][INFO] - Epoch: 60/100
154
+ [2026-07-10 00:47:21,337][vit][INFO] - Train loss=0.5968 acc@1=0.9423 acc@5=0.9878 Time=76.71s
155
+ [2026-07-10 00:47:35,618][vit][INFO] - Val loss=1.9136 acc@1=0.6203 acc@5=0.8434 Time=14.28s
156
+ [2026-07-10 00:47:36,334][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/DeiT-B-16-food101-09-07-26-232741/DeiT-B-16-food101.pt
157
+ [2026-07-10 00:47:36,335][vit][INFO] - Epoch: 61/100
158
+ [2026-07-10 00:48:53,012][vit][INFO] - Train loss=0.5904 acc@1=0.9427 acc@5=0.9881 Time=76.68s
159
+ [2026-07-10 00:48:53,013][vit][INFO] - Epoch: 62/100
160
+ [2026-07-10 00:50:09,705][vit][INFO] - Train loss=0.5879 acc@1=0.9442 acc@5=0.9888 Time=76.69s
161
+ [2026-07-10 00:50:09,705][vit][INFO] - Epoch: 63/100
162
+ [2026-07-10 00:51:26,389][vit][INFO] - Train loss=0.5738 acc@1=0.9466 acc@5=0.9891 Time=76.68s
163
+ [2026-07-10 00:51:26,390][vit][INFO] - Epoch: 64/100
164
+ [2026-07-10 00:52:43,170][vit][INFO] - Train loss=0.5716 acc@1=0.9481 acc@5=0.9901 Time=76.78s
165
+ [2026-07-10 00:52:43,170][vit][INFO] - Epoch: 65/100
166
+ [2026-07-10 00:54:00,108][vit][INFO] - Train loss=0.5563 acc@1=0.9505 acc@5=0.9900 Time=76.94s
167
+ [2026-07-10 00:54:00,109][vit][INFO] - Epoch: 66/100
168
+ [2026-07-10 00:55:17,113][vit][INFO] - Train loss=0.5601 acc@1=0.9507 acc@5=0.9912 Time=77.00s
169
+ [2026-07-10 00:55:17,114][vit][INFO] - Epoch: 67/100
170
+ [2026-07-10 00:56:33,841][vit][INFO] - Train loss=0.5526 acc@1=0.9526 acc@5=0.9910 Time=76.73s
171
+ [2026-07-10 00:56:33,842][vit][INFO] - Epoch: 68/100
172
+ [2026-07-10 00:57:50,541][vit][INFO] - Train loss=0.5549 acc@1=0.9534 acc@5=0.9905 Time=76.70s
173
+ [2026-07-10 00:57:50,542][vit][INFO] - Epoch: 69/100
174
+ [2026-07-10 00:59:07,642][vit][INFO] - Train loss=0.5467 acc@1=0.9550 acc@5=0.9914 Time=77.10s
175
+ [2026-07-10 00:59:07,642][vit][INFO] - Epoch: 70/100
176
+ [2026-07-10 01:00:24,683][vit][INFO] - Train loss=0.5488 acc@1=0.9542 acc@5=0.9912 Time=77.04s
177
+ [2026-07-10 01:00:39,208][vit][INFO] - Val loss=1.9374 acc@1=0.6249 acc@5=0.8442 Time=14.52s
178
+ [2026-07-10 01:00:39,208][vit][INFO] - Epoch: 71/100
179
+ [2026-07-10 01:01:55,956][vit][INFO] - Train loss=0.5384 acc@1=0.9551 acc@5=0.9917 Time=76.75s
180
+ [2026-07-10 01:01:55,957][vit][INFO] - Epoch: 72/100
181
+ [2026-07-10 01:03:12,665][vit][INFO] - Train loss=0.5416 acc@1=0.9563 acc@5=0.9925 Time=76.71s
182
+ [2026-07-10 01:03:12,666][vit][INFO] - Epoch: 73/100
183
+ [2026-07-10 01:04:29,435][vit][INFO] - Train loss=0.5340 acc@1=0.9559 acc@5=0.9920 Time=76.77s
184
+ [2026-07-10 01:04:29,436][vit][INFO] - Epoch: 74/100
185
+ [2026-07-10 01:05:46,283][vit][INFO] - Train loss=0.5388 acc@1=0.9565 acc@5=0.9921 Time=76.85s
186
+ [2026-07-10 01:05:46,284][vit][INFO] - Epoch: 75/100
187
+ [2026-07-10 01:07:03,216][vit][INFO] - Train loss=0.5399 acc@1=0.9575 acc@5=0.9918 Time=76.93s
188
+ [2026-07-10 01:07:03,217][vit][INFO] - Epoch: 76/100
189
+ [2026-07-10 01:08:20,031][vit][INFO] - Train loss=0.5292 acc@1=0.9588 acc@5=0.9923 Time=76.81s
190
+ [2026-07-10 01:08:20,031][vit][INFO] - Epoch: 77/100
191
+ [2026-07-10 01:09:36,861][vit][INFO] - Train loss=0.5293 acc@1=0.9572 acc@5=0.9926 Time=76.83s
192
+ [2026-07-10 01:09:36,861][vit][INFO] - Epoch: 78/100
193
+ [2026-07-10 01:10:53,769][vit][INFO] - Train loss=0.5285 acc@1=0.9579 acc@5=0.9921 Time=76.91s
194
+ [2026-07-10 01:10:53,770][vit][INFO] - Epoch: 79/100
195
+ [2026-07-10 01:12:10,746][vit][INFO] - Train loss=0.5267 acc@1=0.9583 acc@5=0.9929 Time=76.98s
196
+ [2026-07-10 01:12:10,747][vit][INFO] - Epoch: 80/100
197
+ [2026-07-10 01:13:27,901][vit][INFO] - Train loss=0.5254 acc@1=0.9596 acc@5=0.9928 Time=77.15s
198
+ [2026-07-10 01:13:42,004][vit][INFO] - Val loss=1.9679 acc@1=0.6236 acc@5=0.8447 Time=14.10s
199
+ [2026-07-10 01:13:42,664][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/DeiT-B-16-food101-09-07-26-232741/DeiT-B-16-food101.pt
200
+ [2026-07-10 01:13:42,665][vit][INFO] - Epoch: 81/100
201
+ [2026-07-10 01:14:59,422][vit][INFO] - Train loss=0.5225 acc@1=0.9600 acc@5=0.9932 Time=76.76s
202
+ [2026-07-10 01:14:59,424][vit][INFO] - Epoch: 82/100
203
+ [2026-07-10 01:16:16,258][vit][INFO] - Train loss=0.5229 acc@1=0.9589 acc@5=0.9932 Time=76.83s
204
+ [2026-07-10 01:16:16,259][vit][INFO] - Epoch: 83/100
205
+ [2026-07-10 01:17:33,182][vit][INFO] - Train loss=0.5157 acc@1=0.9606 acc@5=0.9935 Time=76.92s
206
+ [2026-07-10 01:17:33,182][vit][INFO] - Epoch: 84/100
207
+ [2026-07-10 01:18:50,039][vit][INFO] - Train loss=0.5178 acc@1=0.9613 acc@5=0.9932 Time=76.86s
208
+ [2026-07-10 01:18:50,040][vit][INFO] - Epoch: 85/100
209
+ [2026-07-10 01:20:07,030][vit][INFO] - Train loss=0.5151 acc@1=0.9603 acc@5=0.9932 Time=76.99s
210
+ [2026-07-10 01:20:07,030][vit][INFO] - Epoch: 86/100
211
+ [2026-07-10 01:21:23,966][vit][INFO] - Train loss=0.5191 acc@1=0.9595 acc@5=0.9933 Time=76.94s
212
+ [2026-07-10 01:21:23,966][vit][INFO] - Epoch: 87/100
213
+ [2026-07-10 01:22:40,836][vit][INFO] - Train loss=0.5152 acc@1=0.9606 acc@5=0.9934 Time=76.87s
214
+ [2026-07-10 01:22:40,837][vit][INFO] - Epoch: 88/100
215
+ [2026-07-10 01:23:57,644][vit][INFO] - Train loss=0.5107 acc@1=0.9616 acc@5=0.9935 Time=76.81s
216
+ [2026-07-10 01:23:57,645][vit][INFO] - Epoch: 89/100
217
+ [2026-07-10 01:25:14,447][vit][INFO] - Train loss=0.5159 acc@1=0.9599 acc@5=0.9935 Time=76.80s
218
+ [2026-07-10 01:25:14,448][vit][INFO] - Epoch: 90/100
219
+ [2026-07-10 01:26:31,342][vit][INFO] - Train loss=0.5090 acc@1=0.9618 acc@5=0.9935 Time=76.89s
220
+ [2026-07-10 01:26:45,898][vit][INFO] - Val loss=1.9708 acc@1=0.6279 acc@5=0.8469 Time=14.55s
221
+ [2026-07-10 01:26:45,899][vit][INFO] - Epoch: 91/100
222
+ [2026-07-10 01:28:02,584][vit][INFO] - Train loss=0.5082 acc@1=0.9623 acc@5=0.9943 Time=76.68s
223
+ [2026-07-10 01:28:02,585][vit][INFO] - Epoch: 92/100
224
+ [2026-07-10 01:29:19,944][vit][INFO] - Train loss=0.5082 acc@1=0.9631 acc@5=0.9941 Time=77.36s
225
+ [2026-07-10 01:29:19,944][vit][INFO] - Epoch: 93/100
226
+ [2026-07-10 01:30:36,619][vit][INFO] - Train loss=0.5065 acc@1=0.9626 acc@5=0.9941 Time=76.67s
227
+ [2026-07-10 01:30:36,620][vit][INFO] - Epoch: 94/100
228
+ [2026-07-10 01:31:53,440][vit][INFO] - Train loss=0.5057 acc@1=0.9628 acc@5=0.9945 Time=76.82s
229
+ [2026-07-10 01:31:53,440][vit][INFO] - Epoch: 95/100
230
+ [2026-07-10 01:33:10,203][vit][INFO] - Train loss=0.5063 acc@1=0.9628 acc@5=0.9940 Time=76.76s
231
+ [2026-07-10 01:33:10,204][vit][INFO] - Epoch: 96/100
232
+ [2026-07-10 01:34:26,987][vit][INFO] - Train loss=0.5037 acc@1=0.9637 acc@5=0.9947 Time=76.78s
233
+ [2026-07-10 01:34:26,988][vit][INFO] - Epoch: 97/100
234
+ [2026-07-10 01:35:43,683][vit][INFO] - Train loss=0.5009 acc@1=0.9641 acc@5=0.9946 Time=76.69s
235
+ [2026-07-10 01:35:43,684][vit][INFO] - Epoch: 98/100
236
+ [2026-07-10 01:37:00,407][vit][INFO] - Train loss=0.5007 acc@1=0.9631 acc@5=0.9941 Time=76.72s
237
+ [2026-07-10 01:37:00,408][vit][INFO] - Epoch: 99/100
238
+ [2026-07-10 01:38:17,188][vit][INFO] - Train loss=0.4990 acc@1=0.9622 acc@5=0.9937 Time=76.78s
239
+ [2026-07-10 01:38:17,189][vit][INFO] - Epoch: 100/100
240
+ [2026-07-10 01:39:34,048][vit][INFO] - Train loss=0.5012 acc@1=0.9634 acc@5=0.9944 Time=76.86s
241
+ [2026-07-10 01:39:48,621][vit][INFO] - Val loss=1.9852 acc@1=0.6293 acc@5=0.8454 Time=14.57s
242
+ [2026-07-10 01:39:49,283][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/DeiT-B-16-food101-09-07-26-232741/DeiT-B-16-food101.pt
DeiT-B-16-food101.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20345a4cc0dbdc8faecfddfb68d8faf4deb58c70e6dfe4128b8edd1c4d46787c
3
+ size 1126870363
README.md ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: pytorch
3
+ pipeline_tag: image-classification
4
+ datasets:
5
+ - food101
6
+ tags:
7
+ - food101
8
+ - image-classification
9
+ - vision-transformer
10
+ - deit
11
+ license: mit
12
+ ---
13
+
14
+ # DeiT-B-16 on Food-101
15
+
16
+ Raw training checkpoint exported from this repository:
17
+ https://github.com/mnjm/vit
18
+
19
+ ## Files
20
+
21
+ - `DeiT-B-16-food101.pt`: raw checkpoint from this training run
22
+ - `config/`: Hydra config snapshot saved with the run
23
+
24
+ ## Checkpoint
25
+
26
+ - Path: `DeiT-B-16-food101.pt`
27
+ - Dataset: `food101`
28
+ - Model: `DeiT-B/16`
29
+ - Teacher: `hf_hub:anonauthors/food101-resnet50`
30
+
31
+ ## Metrics
32
+
33
+ | Split | Loss | Acc@1 | Acc@5 |
34
+ | --- | ---: | ---: | ---: |
35
+ | Train | `0.4401` | `98.16%` | `99.73%` |
36
+ | Val | `1.9847` | `62.92%` | `84.57%` |
37
+
38
+ ## Usage
39
+
40
+ This is a raw checkpoint dump. To use it check [here](https://github.com/mnjm/vit/compute_acc.py)
config/config.yaml ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ dataset:
2
+ name: food101
3
+ img_size: 224
4
+ img_chls: 3
5
+ aug:
6
+ hor_flip_aug: true
7
+ auto_augment: false
8
+ rand_augment: true
9
+ n_class: 101
10
+ model:
11
+ name: DeiT-B-16
12
+ img_size: ${dataset.img_size}
13
+ patch_size: 16
14
+ img_chls: ${dataset.img_chls}
15
+ n_class: ${dataset.n_class}
16
+ n_layer: 12
17
+ n_heads: 12
18
+ n_embd: 768
19
+ mlp_dim: 3072
20
+ drop_rate: 0.1
21
+ attn_drop_rate: 0.1
22
+ stoch_depth_drop_rate: 0.1
23
+ use_dist_token: true
24
+ rng_seed: 31415
25
+ interactive: true
26
+ model_name: ${model.name}-${dataset.name}
27
+ n_epochs: 100
28
+ save_every_epoch: 20
29
+ val_every_epoch: 10
30
+ init_from: scratch
31
+ device_type: cuda
32
+ torch_compile: true
33
+ enable_tf32: true
34
+ clip_grad_norm_1: true
35
+ autocast_dtype: bf16
36
+ wandb: true
37
+ dataloader:
38
+ batch_size: 256
39
+ num_workers: 4
40
+ pin_memory: true
41
+ drop_last: true
42
+ optimizer:
43
+ _target_: torch.optim.AdamW
44
+ lr: 0.0003
45
+ betas:
46
+ - 0.9
47
+ - 0.99
48
+ weight_decay: 0.1
49
+ eps: 1.0e-08
50
+ fused: true
51
+ lr_scheduler:
52
+ name: cosine-with-linear-warmup
53
+ min_lr_pct: 0.1
54
+ warmup_pct: 0.1
55
+ decay_step_pct: 0.6
56
+ deit:
57
+ enable: true
58
+ teacher_name: hf_hub:anonauthors/food101-resnet50
59
+ logging:
60
+ log_dir: ./logs
61
+ wandb:
62
+ enable: ${wandb}
63
+ project: vit
config/hydra.yaml ADDED
@@ -0,0 +1,159 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: ${logging.log_dir}/${hydra.job.name}
4
+ sweep:
5
+ dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
6
+ subdir: ${hydra.job.num}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ formatters:
71
+ simple:
72
+ format: '[%(asctime)s][HYDRA] %(message)s'
73
+ handlers:
74
+ console:
75
+ class: logging.StreamHandler
76
+ formatter: simple
77
+ stream: ext://sys.stdout
78
+ root:
79
+ level: INFO
80
+ handlers:
81
+ - console
82
+ loggers:
83
+ logging_example:
84
+ level: DEBUG
85
+ disable_existing_loggers: false
86
+ job_logging:
87
+ version: 1
88
+ formatters:
89
+ simple:
90
+ format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
91
+ handlers:
92
+ console:
93
+ class: logging.StreamHandler
94
+ formatter: simple
95
+ stream: ext://sys.stdout
96
+ file:
97
+ class: logging.FileHandler
98
+ formatter: simple
99
+ filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
100
+ root:
101
+ level: INFO
102
+ handlers:
103
+ - console
104
+ - file
105
+ disable_existing_loggers: false
106
+ env: {}
107
+ mode: RUN
108
+ searchpath: []
109
+ callbacks: {}
110
+ output_subdir: config
111
+ overrides:
112
+ hydra:
113
+ - hydra.mode=RUN
114
+ task:
115
+ - +run=deit-food101
116
+ - wandb=true
117
+ job:
118
+ name: ${model_name}-${now_ist:%d-%m-%y-%H%M%S}
119
+ chdir: null
120
+ override_dirname: +run=deit-food101,wandb=true
121
+ id: ???
122
+ num: ???
123
+ config_name: default
124
+ env_set: {}
125
+ env_copy: []
126
+ config:
127
+ override_dirname:
128
+ kv_sep: '='
129
+ item_sep: ','
130
+ exclude_keys: []
131
+ runtime:
132
+ version: 1.3.2
133
+ version_base: '1.3'
134
+ cwd: /home/noshai/workspace/scratch-pad/modelz-esp
135
+ config_sources:
136
+ - path: hydra.conf
137
+ schema: pkg
138
+ provider: hydra
139
+ - path: /home/noshai/workspace/scratch-pad/modelz-esp/config
140
+ schema: file
141
+ provider: main
142
+ - path: ''
143
+ schema: structured
144
+ provider: schema
145
+ output_dir: /home/noshai/workspace/scratch-pad/modelz-esp/logs/DeiT-B-16-food101-09-07-26-232741
146
+ choices:
147
+ run: deit-food101
148
+ model: DeiT-B-16
149
+ dataset: food101
150
+ hydra/env: default
151
+ hydra/callbacks: null
152
+ hydra/job_logging: default
153
+ hydra/hydra_logging: default
154
+ hydra/hydra_help: default
155
+ hydra/help: default
156
+ hydra/sweeper: basic
157
+ hydra/launcher: basic
158
+ hydra/output: default
159
+ verbose: false
config/overrides.yaml ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ - +run=deit-food101
2
+ - wandb=true