Upload folder using huggingface_hub

README.md

@@ -0,0 +1,39 @@
+---
+library_name: pytorch
+pipeline_tag: image-classification
+datasets:
+- food101
+tags:
+- food101
+- image-classification
+- vision-transformer
+- vit
+license: mit
+---
+
+# ViT-B-16 on Food-101
+
+Raw training checkpoint exported from this repository:
+https://github.com/mnjm/vit
+
+## Files
+
+- `ViT-B-16-food101.pt`: raw checkpoint from this training run
+- `config/`: Hydra config snapshot saved with the run
+
+## Checkpoint
+
+- Path: `ViT-B-16-food101.pt`
+- Dataset: `food101`
+- Model: `ViT-B/16`
+
+## Metrics
+
+| Split | Loss | Acc@1 | Acc@5 |
+| --- | ---: | ---: | ---: |
+| Train | `0.0227` | `99.41%` | `99.87%` |
+| Val | `2.5183` | `58.57%` | `81.41%` |
+
+## Usage
+
+This is a raw checkpoint dump. To use it check [here](https://github.com/mnjm/vit/compute_acc.py)

ViT-B-16-food101.log

@@ -0,0 +1,236 @@
+[2026-07-09 21:33:08,395][vit][INFO] - Using cuda
+[2026-07-09 21:33:08,396][vit][INFO] - Loading food101 dataset
+[2026-07-09 21:33:08,724][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
+[2026-07-09 21:33:08,732][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/ethz/food101/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/README.md "HTTP/1.1 200 OK"
+[2026-07-09 21:33:08,964][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/food101.py "HTTP/1.1 404 Not Found"
+[2026-07-09 21:33:09,677][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"
+[2026-07-09 21:33:09,911][httpx][INFO] - HTTP Request: GET https://huggingface.co/api/datasets/ethz/food101/revision/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9 "HTTP/1.1 200 OK"
+[2026-07-09 21:33:10,146][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/.huggingface.yaml "HTTP/1.1 404 Not Found"
+[2026-07-09 21:33:10,471][httpx][INFO] - HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=ethz/food101 "HTTP/1.1 200 OK"
+[2026-07-09 21:33:10,709][httpx][INFO] - HTTP Request: GET https://huggingface.co/api/datasets/ethz/food101/tree/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/data?recursive=true&expand=false "HTTP/1.1 200 OK"
+[2026-07-09 21:33:10,946][httpx][INFO] - HTTP Request: GET https://huggingface.co/api/datasets/ethz/food101/tree/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9?recursive=false&expand=false "HTTP/1.1 200 OK"
+[2026-07-09 21:33:11,177][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/dataset_infos.json "HTTP/1.1 404 Not Found"
+[2026-07-09 21:33:11,432][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/main/README.md "HTTP/1.1 307 Temporary Redirect"
+[2026-07-09 21:33:11,444][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/api/resolve-cache/datasets/ethz/food101/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/README.md "HTTP/1.1 200 OK"
+[2026-07-09 21:33:11,679][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/food101.py "HTTP/1.1 404 Not Found"
+[2026-07-09 21:33:11,903][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"
+[2026-07-09 21:33:12,144][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/.huggingface.yaml "HTTP/1.1 404 Not Found"
+[2026-07-09 21:33:12,386][httpx][INFO] - HTTP Request: GET https://datasets-server.huggingface.co/info?dataset=ethz/food101 "HTTP/1.1 200 OK"
+[2026-07-09 21:33:12,624][httpx][INFO] - HTTP Request: HEAD https://huggingface.co/datasets/ethz/food101/resolve/83488de741c1bd1ce27aa6a2b33e19c7bdf92ca9/dataset_infos.json "HTTP/1.1 404 Not Found"
+[2026-07-09 21:33:14,635][vit][INFO] - Model type: ViT-B-16 params: 85,876,325
+[2026-07-09 21:33:33,869][vit][INFO] - Initial Val loss=4.7851 acc@1=0.0095 acc@5=0.0470 Time=15.80s
+[2026-07-09 21:33:33,870][vit][INFO] - Epoch: 1/100
+[2026-07-09 21:34:29,024][vit][INFO] - Train loss=4.5911 acc@1=0.0211 acc@5=0.0868 Time=55.15s
+[2026-07-09 21:34:29,024][vit][INFO] - Epoch: 2/100
+[2026-07-09 21:35:35,278][vit][INFO] - Train loss=4.3031 acc@1=0.0552 acc@5=0.1829 Time=66.25s
+[2026-07-09 21:35:35,278][vit][INFO] - Epoch: 3/100
+[2026-07-09 21:36:41,569][vit][INFO] - Train loss=4.0183 acc@1=0.0951 acc@5=0.2734 Time=66.29s
+[2026-07-09 21:36:41,570][vit][INFO] - Epoch: 4/100
+[2026-07-09 21:37:47,780][vit][INFO] - Train loss=3.8221 acc@1=0.1252 acc@5=0.3321 Time=66.21s
+[2026-07-09 21:37:47,780][vit][INFO] - Epoch: 5/100
+[2026-07-09 21:38:53,995][vit][INFO] - Train loss=3.6753 acc@1=0.1512 acc@5=0.3732 Time=66.21s
+[2026-07-09 21:38:53,995][vit][INFO] - Epoch: 6/100
+[2026-07-09 21:40:00,359][vit][INFO] - Train loss=3.5549 acc@1=0.1698 acc@5=0.4070 Time=66.36s
+[2026-07-09 21:40:00,359][vit][INFO] - Epoch: 7/100
+[2026-07-09 21:41:06,988][vit][INFO] - Train loss=3.4547 acc@1=0.1910 acc@5=0.4330 Time=66.63s
+[2026-07-09 21:41:06,989][vit][INFO] - Epoch: 8/100
+[2026-07-09 21:42:13,768][vit][INFO] - Train loss=3.3529 acc@1=0.2067 acc@5=0.4613 Time=66.78s
+[2026-07-09 21:42:13,769][vit][INFO] - Epoch: 9/100
+[2026-07-09 21:43:20,312][vit][INFO] - Train loss=3.2805 acc@1=0.2212 acc@5=0.4811 Time=66.54s
+[2026-07-09 21:43:20,312][vit][INFO] - Epoch: 10/100
+[2026-07-09 21:44:26,942][vit][INFO] - Train loss=3.2206 acc@1=0.2330 acc@5=0.4951 Time=66.63s
+[2026-07-09 21:44:40,895][vit][INFO] - Val   loss=2.7674 acc@1=0.3159 acc@5=0.6061 Time=13.95s
+[2026-07-09 21:44:40,897][vit][INFO] - Epoch: 11/100
+[2026-07-09 21:45:47,285][vit][INFO] - Train loss=3.1368 acc@1=0.2489 acc@5=0.5157 Time=66.39s
+[2026-07-09 21:45:47,285][vit][INFO] - Epoch: 12/100
+[2026-07-09 21:46:53,757][vit][INFO] - Train loss=3.0434 acc@1=0.2688 acc@5=0.5382 Time=66.47s
+[2026-07-09 21:46:53,758][vit][INFO] - Epoch: 13/100
+[2026-07-09 21:48:00,223][vit][INFO] - Train loss=2.9338 acc@1=0.2895 acc@5=0.5662 Time=66.47s
+[2026-07-09 21:48:00,225][vit][INFO] - Epoch: 14/100
+[2026-07-09 21:49:07,030][vit][INFO] - Train loss=2.8407 acc@1=0.3086 acc@5=0.5880 Time=66.81s
+[2026-07-09 21:49:07,031][vit][INFO] - Epoch: 15/100
+[2026-07-09 21:50:13,599][vit][INFO] - Train loss=2.7408 acc@1=0.3273 acc@5=0.6093 Time=66.57s
+[2026-07-09 21:50:13,599][vit][INFO] - Epoch: 16/100
+[2026-07-09 21:51:20,220][vit][INFO] - Train loss=2.6490 acc@1=0.3478 acc@5=0.6325 Time=66.62s
+[2026-07-09 21:51:20,221][vit][INFO] - Epoch: 17/100
+[2026-07-09 21:52:26,931][vit][INFO] - Train loss=2.5627 acc@1=0.3654 acc@5=0.6499 Time=66.71s
+[2026-07-09 21:52:26,932][vit][INFO] - Epoch: 18/100
+[2026-07-09 21:53:33,401][vit][INFO] - Train loss=2.4682 acc@1=0.3857 acc@5=0.6685 Time=66.47s
+[2026-07-09 21:53:33,402][vit][INFO] - Epoch: 19/100
+[2026-07-09 21:54:39,839][vit][INFO] - Train loss=2.3714 acc@1=0.4049 acc@5=0.6882 Time=66.44s
+[2026-07-09 21:54:39,840][vit][INFO] - Epoch: 20/100
+[2026-07-09 21:55:46,328][vit][INFO] - Train loss=2.2842 acc@1=0.4230 acc@5=0.7078 Time=66.49s
+[2026-07-09 21:56:00,450][vit][INFO] - Val   loss=1.9792 acc@1=0.4906 acc@5=0.7670 Time=14.12s
+[2026-07-09 21:56:00,853][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/ViT-B-16-food101-09-07-26-213308/ViT-B-16-food101.pt
+[2026-07-09 21:56:00,853][vit][INFO] - Epoch: 21/100
+[2026-07-09 21:57:07,319][vit][INFO] - Train loss=2.1950 acc@1=0.4396 acc@5=0.7237 Time=66.46s
+[2026-07-09 21:57:07,319][vit][INFO] - Epoch: 22/100
+[2026-07-09 21:58:13,871][vit][INFO] - Train loss=2.1014 acc@1=0.4619 acc@5=0.7414 Time=66.55s
+[2026-07-09 21:58:13,871][vit][INFO] - Epoch: 23/100
+[2026-07-09 21:59:20,386][vit][INFO] - Train loss=2.0076 acc@1=0.4820 acc@5=0.7601 Time=66.51s
+[2026-07-09 21:59:20,387][vit][INFO] - Epoch: 24/100
+[2026-07-09 22:00:26,939][vit][INFO] - Train loss=1.9141 acc@1=0.5020 acc@5=0.7777 Time=66.55s
+[2026-07-09 22:00:26,940][vit][INFO] - Epoch: 25/100
+[2026-07-09 22:01:33,604][vit][INFO] - Train loss=1.8151 acc@1=0.5247 acc@5=0.7950 Time=66.66s
+[2026-07-09 22:01:33,605][vit][INFO] - Epoch: 26/100
+[2026-07-09 22:02:40,064][vit][INFO] - Train loss=1.7139 acc@1=0.5466 acc@5=0.8130 Time=66.46s
+[2026-07-09 22:02:40,064][vit][INFO] - Epoch: 27/100
+[2026-07-09 22:03:46,562][vit][INFO] - Train loss=1.6133 acc@1=0.5706 acc@5=0.8297 Time=66.50s
+[2026-07-09 22:03:46,562][vit][INFO] - Epoch: 28/100
+[2026-07-09 22:04:53,276][vit][INFO] - Train loss=1.5212 acc@1=0.5913 acc@5=0.8454 Time=66.71s
+[2026-07-09 22:04:53,276][vit][INFO] - Epoch: 29/100
+[2026-07-09 22:05:59,881][vit][INFO] - Train loss=1.4165 acc@1=0.6170 acc@5=0.8626 Time=66.60s
+[2026-07-09 22:05:59,881][vit][INFO] - Epoch: 30/100
+[2026-07-09 22:07:06,323][vit][INFO] - Train loss=1.3241 acc@1=0.6402 acc@5=0.8762 Time=66.44s
+[2026-07-09 22:07:20,316][vit][INFO] - Val   loss=1.8612 acc@1=0.5469 acc@5=0.8050 Time=13.99s
+[2026-07-09 22:07:20,317][vit][INFO] - Epoch: 31/100
+[2026-07-09 22:08:26,532][vit][INFO] - Train loss=1.2366 acc@1=0.6618 acc@5=0.8890 Time=66.21s
+[2026-07-09 22:08:26,532][vit][INFO] - Epoch: 32/100
+[2026-07-09 22:09:32,950][vit][INFO] - Train loss=1.1418 acc@1=0.6841 acc@5=0.9033 Time=66.42s
+[2026-07-09 22:09:32,951][vit][INFO] - Epoch: 33/100
+[2026-07-09 22:10:39,636][vit][INFO] - Train loss=1.0586 acc@1=0.7063 acc@5=0.9140 Time=66.69s
+[2026-07-09 22:10:39,636][vit][INFO] - Epoch: 34/100
+[2026-07-09 22:11:46,747][vit][INFO] - Train loss=0.9811 acc@1=0.7279 acc@5=0.9225 Time=67.11s
+[2026-07-09 22:11:46,747][vit][INFO] - Epoch: 35/100
+[2026-07-09 22:12:53,310][vit][INFO] - Train loss=0.9052 acc@1=0.7477 acc@5=0.9318 Time=66.56s
+[2026-07-09 22:12:53,310][vit][INFO] - Epoch: 36/100
+[2026-07-09 22:13:59,856][vit][INFO] - Train loss=0.8385 acc@1=0.7652 acc@5=0.9390 Time=66.55s
+[2026-07-09 22:13:59,857][vit][INFO] - Epoch: 37/100
+[2026-07-09 22:15:06,516][vit][INFO] - Train loss=0.7765 acc@1=0.7824 acc@5=0.9454 Time=66.66s
+[2026-07-09 22:15:06,516][vit][INFO] - Epoch: 38/100
+[2026-07-09 22:16:13,244][vit][INFO] - Train loss=0.7169 acc@1=0.7972 acc@5=0.9515 Time=66.73s
+[2026-07-09 22:16:13,246][vit][INFO] - Epoch: 39/100
+[2026-07-09 22:17:20,058][vit][INFO] - Train loss=0.6647 acc@1=0.8119 acc@5=0.9575 Time=66.81s
+[2026-07-09 22:17:20,058][vit][INFO] - Epoch: 40/100
+[2026-07-09 22:18:27,076][vit][INFO] - Train loss=0.6283 acc@1=0.8232 acc@5=0.9597 Time=67.02s
+[2026-07-09 22:18:41,270][vit][INFO] - Val   loss=2.1153 acc@1=0.5580 acc@5=0.8027 Time=14.19s
+[2026-07-09 22:18:41,876][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/ViT-B-16-food101-09-07-26-213308/ViT-B-16-food101.pt
+[2026-07-09 22:18:41,876][vit][INFO] - Epoch: 41/100
+[2026-07-09 22:19:48,947][vit][INFO] - Train loss=0.5749 acc@1=0.8367 acc@5=0.9640 Time=67.07s
+[2026-07-09 22:19:48,948][vit][INFO] - Epoch: 42/100
+[2026-07-09 22:20:56,220][vit][INFO] - Train loss=0.5427 acc@1=0.8471 acc@5=0.9662 Time=67.27s
+[2026-07-09 22:20:56,220][vit][INFO] - Epoch: 43/100
+[2026-07-09 22:22:03,608][vit][INFO] - Train loss=0.5017 acc@1=0.8584 acc@5=0.9697 Time=67.39s
+[2026-07-09 22:22:03,609][vit][INFO] - Epoch: 44/100
+[2026-07-09 22:23:11,057][vit][INFO] - Train loss=0.4723 acc@1=0.8673 acc@5=0.9721 Time=67.45s
+[2026-07-09 22:23:11,057][vit][INFO] - Epoch: 45/100
+[2026-07-09 22:24:18,555][vit][INFO] - Train loss=0.4379 acc@1=0.8746 acc@5=0.9757 Time=67.50s
+[2026-07-09 22:24:18,556][vit][INFO] - Epoch: 46/100
+[2026-07-09 22:25:25,964][vit][INFO] - Train loss=0.4092 acc@1=0.8846 acc@5=0.9767 Time=67.41s
+[2026-07-09 22:25:25,965][vit][INFO] - Epoch: 47/100
+[2026-07-09 22:26:33,368][vit][INFO] - Train loss=0.3812 acc@1=0.8929 acc@5=0.9798 Time=67.40s
+[2026-07-09 22:26:33,369][vit][INFO] - Epoch: 48/100
+[2026-07-09 22:27:40,887][vit][INFO] - Train loss=0.3620 acc@1=0.8983 acc@5=0.9800 Time=67.52s
+[2026-07-09 22:27:40,887][vit][INFO] - Epoch: 49/100
+[2026-07-09 22:28:48,360][vit][INFO] - Train loss=0.3375 acc@1=0.9049 acc@5=0.9816 Time=67.47s
+[2026-07-09 22:28:48,360][vit][INFO] - Epoch: 50/100
+[2026-07-09 22:29:55,984][vit][INFO] - Train loss=0.3183 acc@1=0.9107 acc@5=0.9827 Time=67.62s
+[2026-07-09 22:30:09,722][vit][INFO] - Val   loss=2.2642 acc@1=0.5711 acc@5=0.8099 Time=13.74s
+[2026-07-09 22:30:09,723][vit][INFO] - Epoch: 51/100
+[2026-07-09 22:31:16,956][vit][INFO] - Train loss=0.2955 acc@1=0.9177 acc@5=0.9848 Time=67.23s
+[2026-07-09 22:31:16,957][vit][INFO] - Epoch: 52/100
+[2026-07-09 22:32:24,424][vit][INFO] - Train loss=0.2801 acc@1=0.9218 acc@5=0.9851 Time=67.47s
+[2026-07-09 22:32:24,425][vit][INFO] - Epoch: 53/100
+[2026-07-09 22:33:31,884][vit][INFO] - Train loss=0.2609 acc@1=0.9269 acc@5=0.9867 Time=67.46s
+[2026-07-09 22:33:31,885][vit][INFO] - Epoch: 54/100
+[2026-07-09 22:34:39,212][vit][INFO] - Train loss=0.2470 acc@1=0.9311 acc@5=0.9872 Time=67.33s
+[2026-07-09 22:34:39,212][vit][INFO] - Epoch: 55/100
+[2026-07-09 22:35:46,538][vit][INFO] - Train loss=0.2332 acc@1=0.9346 acc@5=0.9886 Time=67.33s
+[2026-07-09 22:35:46,538][vit][INFO] - Epoch: 56/100
+[2026-07-09 22:36:54,125][vit][INFO] - Train loss=0.2170 acc@1=0.9386 acc@5=0.9895 Time=67.59s
+[2026-07-09 22:36:54,125][vit][INFO] - Epoch: 57/100
+[2026-07-09 22:38:01,647][vit][INFO] - Train loss=0.2013 acc@1=0.9430 acc@5=0.9900 Time=67.52s
+[2026-07-09 22:38:01,647][vit][INFO] - Epoch: 58/100
+[2026-07-09 22:39:09,108][vit][INFO] - Train loss=0.1932 acc@1=0.9456 acc@5=0.9910 Time=67.46s
+[2026-07-09 22:39:09,109][vit][INFO] - Epoch: 59/100
+[2026-07-09 22:40:16,731][vit][INFO] - Train loss=0.1805 acc@1=0.9504 acc@5=0.9912 Time=67.62s
+[2026-07-09 22:40:16,732][vit][INFO] - Epoch: 60/100
+[2026-07-09 22:41:24,207][vit][INFO] - Train loss=0.1748 acc@1=0.9518 acc@5=0.9915 Time=67.48s
+[2026-07-09 22:41:38,132][vit][INFO] - Val   loss=2.3022 acc@1=0.5847 acc@5=0.8155 Time=13.92s
+[2026-07-09 22:41:38,738][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/ViT-B-16-food101-09-07-26-213308/ViT-B-16-food101.pt
+[2026-07-09 22:41:38,738][vit][INFO] - Epoch: 61/100
+[2026-07-09 22:42:45,957][vit][INFO] - Train loss=0.1588 acc@1=0.9556 acc@5=0.9926 Time=67.22s
+[2026-07-09 22:42:45,957][vit][INFO] - Epoch: 62/100
+[2026-07-09 22:43:53,330][vit][INFO] - Train loss=0.1548 acc@1=0.9578 acc@5=0.9926 Time=67.37s
+[2026-07-09 22:43:53,331][vit][INFO] - Epoch: 63/100
+[2026-07-09 22:45:00,770][vit][INFO] - Train loss=0.1492 acc@1=0.9588 acc@5=0.9928 Time=67.44s
+[2026-07-09 22:45:00,770][vit][INFO] - Epoch: 64/100
+[2026-07-09 22:46:08,205][vit][INFO] - Train loss=0.1423 acc@1=0.9605 acc@5=0.9934 Time=67.43s
+[2026-07-09 22:46:08,206][vit][INFO] - Epoch: 65/100
+[2026-07-09 22:47:15,614][vit][INFO] - Train loss=0.1335 acc@1=0.9623 acc@5=0.9940 Time=67.41s
+[2026-07-09 22:47:15,615][vit][INFO] - Epoch: 66/100
+[2026-07-09 22:48:22,998][vit][INFO] - Train loss=0.1327 acc@1=0.9634 acc@5=0.9938 Time=67.38s
+[2026-07-09 22:48:22,998][vit][INFO] - Epoch: 67/100
+[2026-07-09 22:49:30,591][vit][INFO] - Train loss=0.1257 acc@1=0.9653 acc@5=0.9940 Time=67.59s
+[2026-07-09 22:49:30,592][vit][INFO] - Epoch: 68/100
+[2026-07-09 22:50:37,975][vit][INFO] - Train loss=0.1245 acc@1=0.9654 acc@5=0.9942 Time=67.38s
+[2026-07-09 22:50:37,977][vit][INFO] - Epoch: 69/100
+[2026-07-09 22:51:45,470][vit][INFO] - Train loss=0.1198 acc@1=0.9675 acc@5=0.9944 Time=67.49s
+[2026-07-09 22:51:45,470][vit][INFO] - Epoch: 70/100
+[2026-07-09 22:52:52,922][vit][INFO] - Train loss=0.1148 acc@1=0.9681 acc@5=0.9951 Time=67.45s
+[2026-07-09 22:53:06,631][vit][INFO] - Val   loss=2.3806 acc@1=0.5885 acc@5=0.8139 Time=13.71s
+[2026-07-09 22:53:06,632][vit][INFO] - Epoch: 71/100
+[2026-07-09 22:54:13,941][vit][INFO] - Train loss=0.1168 acc@1=0.9681 acc@5=0.9944 Time=67.31s
+[2026-07-09 22:54:13,942][vit][INFO] - Epoch: 72/100
+[2026-07-09 22:55:21,259][vit][INFO] - Train loss=0.1103 acc@1=0.9698 acc@5=0.9954 Time=67.32s
+[2026-07-09 22:55:21,260][vit][INFO] - Epoch: 73/100
+[2026-07-09 22:56:28,717][vit][INFO] - Train loss=0.1108 acc@1=0.9691 acc@5=0.9951 Time=67.46s
+[2026-07-09 22:56:28,718][vit][INFO] - Epoch: 74/100
+[2026-07-09 22:57:36,022][vit][INFO] - Train loss=0.1094 acc@1=0.9704 acc@5=0.9952 Time=67.30s
+[2026-07-09 22:57:36,022][vit][INFO] - Epoch: 75/100
+[2026-07-09 22:58:43,479][vit][INFO] - Train loss=0.1099 acc@1=0.9693 acc@5=0.9950 Time=67.46s
+[2026-07-09 22:58:43,480][vit][INFO] - Epoch: 76/100
+[2026-07-09 22:59:51,120][vit][INFO] - Train loss=0.1065 acc@1=0.9705 acc@5=0.9954 Time=67.64s
+[2026-07-09 22:59:51,120][vit][INFO] - Epoch: 77/100
+[2026-07-09 23:00:58,502][vit][INFO] - Train loss=0.1031 acc@1=0.9714 acc@5=0.9956 Time=67.38s
+[2026-07-09 23:00:58,503][vit][INFO] - Epoch: 78/100
+[2026-07-09 23:02:05,944][vit][INFO] - Train loss=0.1022 acc@1=0.9713 acc@5=0.9959 Time=67.44s
+[2026-07-09 23:02:05,944][vit][INFO] - Epoch: 79/100
+[2026-07-09 23:03:13,351][vit][INFO] - Train loss=0.1006 acc@1=0.9727 acc@5=0.9958 Time=67.41s
+[2026-07-09 23:03:13,352][vit][INFO] - Epoch: 80/100
+[2026-07-09 23:04:20,675][vit][INFO] - Train loss=0.0985 acc@1=0.9733 acc@5=0.9953 Time=67.32s
+[2026-07-09 23:04:34,528][vit][INFO] - Val   loss=2.4279 acc@1=0.5872 acc@5=0.8152 Time=13.85s
+[2026-07-09 23:04:35,122][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/ViT-B-16-food101-09-07-26-213308/ViT-B-16-food101.pt
+[2026-07-09 23:04:35,122][vit][INFO] - Epoch: 81/100
+[2026-07-09 23:05:42,273][vit][INFO] - Train loss=0.0996 acc@1=0.9721 acc@5=0.9956 Time=67.15s
+[2026-07-09 23:05:42,273][vit][INFO] - Epoch: 82/100
+[2026-07-09 23:06:49,698][vit][INFO] - Train loss=0.0974 acc@1=0.9725 acc@5=0.9961 Time=67.42s
+[2026-07-09 23:06:49,698][vit][INFO] - Epoch: 83/100
+[2026-07-09 23:07:57,146][vit][INFO] - Train loss=0.0988 acc@1=0.9727 acc@5=0.9954 Time=67.45s
+[2026-07-09 23:07:57,147][vit][INFO] - Epoch: 84/100
+[2026-07-09 23:09:04,790][vit][INFO] - Train loss=0.0953 acc@1=0.9737 acc@5=0.9962 Time=67.64s
+[2026-07-09 23:09:04,790][vit][INFO] - Epoch: 85/100
+[2026-07-09 23:10:12,193][vit][INFO] - Train loss=0.0927 acc@1=0.9741 acc@5=0.9960 Time=67.40s
+[2026-07-09 23:10:12,194][vit][INFO] - Epoch: 86/100
+[2026-07-09 23:11:19,615][vit][INFO] - Train loss=0.0962 acc@1=0.9732 acc@5=0.9961 Time=67.42s
+[2026-07-09 23:11:19,616][vit][INFO] - Epoch: 87/100
+[2026-07-09 23:12:27,090][vit][INFO] - Train loss=0.0921 acc@1=0.9743 acc@5=0.9960 Time=67.47s
+[2026-07-09 23:12:27,091][vit][INFO] - Epoch: 88/100
+[2026-07-09 23:13:34,521][vit][INFO] - Train loss=0.0892 acc@1=0.9750 acc@5=0.9963 Time=67.43s
+[2026-07-09 23:13:34,522][vit][INFO] - Epoch: 89/100
+[2026-07-09 23:14:41,826][vit][INFO] - Train loss=0.0891 acc@1=0.9748 acc@5=0.9964 Time=67.30s
+[2026-07-09 23:14:41,827][vit][INFO] - Epoch: 90/100
+[2026-07-09 23:15:49,301][vit][INFO] - Train loss=0.0925 acc@1=0.9740 acc@5=0.9961 Time=67.47s
+[2026-07-09 23:16:03,288][vit][INFO] - Val   loss=2.4194 acc@1=0.5917 acc@5=0.8164 Time=13.99s
+[2026-07-09 23:16:03,289][vit][INFO] - Epoch: 91/100
+[2026-07-09 23:17:10,420][vit][INFO] - Train loss=0.0922 acc@1=0.9742 acc@5=0.9961 Time=67.13s
+[2026-07-09 23:17:10,420][vit][INFO] - Epoch: 92/100
+[2026-07-09 23:18:17,697][vit][INFO] - Train loss=0.0859 acc@1=0.9764 acc@5=0.9967 Time=67.28s
+[2026-07-09 23:18:17,698][vit][INFO] - Epoch: 93/100
+[2026-07-09 23:19:25,177][vit][INFO] - Train loss=0.0860 acc@1=0.9762 acc@5=0.9966 Time=67.48s
+[2026-07-09 23:19:25,178][vit][INFO] - Epoch: 94/100
+[2026-07-09 23:20:32,600][vit][INFO] - Train loss=0.0849 acc@1=0.9765 acc@5=0.9962 Time=67.42s
+[2026-07-09 23:20:32,600][vit][INFO] - Epoch: 95/100
+[2026-07-09 23:21:40,037][vit][INFO] - Train loss=0.0836 acc@1=0.9771 acc@5=0.9966 Time=67.44s
+[2026-07-09 23:21:40,038][vit][INFO] - Epoch: 96/100
+[2026-07-09 23:22:47,410][vit][INFO] - Train loss=0.0847 acc@1=0.9767 acc@5=0.9966 Time=67.37s
+[2026-07-09 23:22:47,411][vit][INFO] - Epoch: 97/100
+[2026-07-09 23:23:54,876][vit][INFO] - Train loss=0.0814 acc@1=0.9776 acc@5=0.9964 Time=67.46s
+[2026-07-09 23:23:54,877][vit][INFO] - Epoch: 98/100
+[2026-07-09 23:25:02,329][vit][INFO] - Train loss=0.0856 acc@1=0.9766 acc@5=0.9967 Time=67.45s
+[2026-07-09 23:25:02,330][vit][INFO] - Epoch: 99/100
+[2026-07-09 23:26:09,747][vit][INFO] - Train loss=0.0812 acc@1=0.9774 acc@5=0.9968 Time=67.42s
+[2026-07-09 23:26:09,748][vit][INFO] - Epoch: 100/100
+[2026-07-09 23:27:17,089][vit][INFO] - Train loss=0.0826 acc@1=0.9774 acc@5=0.9966 Time=67.34s
+[2026-07-09 23:27:30,873][vit][INFO] - Val   loss=2.5193 acc@1=0.5857 acc@5=0.8140 Time=13.78s
+[2026-07-09 23:27:31,508][vit][INFO] - Saved checkpoint to /home/noshai/workspace/scratch-pad/modelz-esp/logs/ViT-B-16-food101-09-07-26-213308/ViT-B-16-food101.pt

ViT-B-16-food101.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccd90ce91d16aabe1fd85654160252dfb764aaf77c049e229c74c1b37c669b3f
+size 1030722389

config/config.yaml

@@ -0,0 +1,62 @@
+dataset:
+  name: food101
+  img_size: 224
+  img_chls: 3
+  aug:
+    hor_flip_aug: true
+    auto_augment: false
+    rand_augment: true
+  n_class: 101
+model:
+  name: ViT-B-16
+  img_size: ${dataset.img_size}
+  patch_size: 16
+  img_chls: ${dataset.img_chls}
+  n_class: ${dataset.n_class}
+  n_layer: 12
+  n_heads: 12
+  n_embd: 768
+  mlp_dim: 3072
+  drop_rate: 0.1
+  attn_drop_rate: 0.1
+  stoch_depth_drop_rate: 0.2
+rng_seed: 31415
+interactive: true
+model_name: ${model.name}-${dataset.name}
+n_epochs: 100
+save_every_epoch: 20
+val_every_epoch: 10
+init_from: scratch
+device_type: cuda
+torch_compile: true
+enable_tf32: true
+clip_grad_norm_1: true
+autocast_dtype: bf16
+wandb: true
+dataloader:
+  batch_size: 256
+  num_workers: 4
+  pin_memory: true
+  drop_last: true
+optimizer:
+  _target_: torch.optim.AdamW
+  lr: 0.0003
+  betas:
+  - 0.9
+  - 0.99
+  weight_decay: 0.1
+  eps: 1.0e-08
+  fused: true
+lr_scheduler:
+  name: cosine-with-linear-warmup
+  min_lr_pct: 0.1
+  warmup_pct: 0.1
+  decay_step_pct: 0.6
+deit:
+  enable: false
+  teacher_name: null
+logging:
+  log_dir: ./logs
+  wandb:
+    enable: ${wandb}
+    project: vit

config/hydra.yaml

@@ -0,0 +1,159 @@
+hydra:
+  run:
+    dir: ${logging.log_dir}/${hydra.job.name}
+  sweep:
+    dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: config
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task:
+    - +run=vit-food101
+    - wandb=true
+  job:
+    name: ${model_name}-${now_ist:%d-%m-%y-%H%M%S}
+    chdir: null
+    override_dirname: +run=vit-food101,wandb=true
+    id: ???
+    num: ???
+    config_name: default
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.3.2
+    version_base: '1.3'
+    cwd: /home/noshai/workspace/scratch-pad/modelz-esp
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /home/noshai/workspace/scratch-pad/modelz-esp/config
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /home/noshai/workspace/scratch-pad/modelz-esp/logs/ViT-B-16-food101-09-07-26-213308
+    choices:
+      run: vit-food101
+      model: ViT-B-16
+      dataset: food101
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

config/overrides.yaml

@@ -0,0 +1,2 @@
+- +run=vit-food101
+- wandb=true