Downloads · 30 days
0
capit-sat
Bukunmi2108/capit-sat
capit-sat is a image-to-text model from Bukunmi2108. Use it when you need a caption or text from an image. It is set up for pytorch. The card lists the license as mit.
Show, Attend and Tell image captioner, trained from scratch on Flickr8k (Karpathy split). The glass-box half of capit — exposes per-word attention, beam candidates, and word-by-word playback.
Access
Public
Updated Jun 10, 2026
Repo size
148 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt148 MB · 100%
At a glance
- Task
- Image-to-Text
- Library
- pytorch
- License
- mit
- Access
- Public
- Created
- Jun 10, 2026
- Updated
- Jun 10, 2026
- SHA
- 47662d21
Datasets
Spaces
Model card
From the Hugging Face model README
capit-sat
Show, Attend and Tell image captioner, trained from scratch on Flickr8k (Karpathy split). The glass-box half of capit — exposes per-word attention, beam candidates, and word-by-word playback.
Test-set scores (pycocoevalcap, Karpathy test = 1000 images)
| beam | BLEU-1 | BLEU-2 | BLEU-3 | BLEU-4 | CIDEr |
|---|---|---|---|---|---|
| 1 | 61.99 | 44.37 | 30.23 | 20.05 | 55.51 |
| 3 | 64.77 | 47.34 | 33.68 | 23.45 | 62.20 |
| 5 | 65.54 | 47.84 | 34.08 | 23.63 | 62.80 |
Training
- Backbone: frozen ResNet-50 (ImageNet). Decoder trained from scratch.
- Best val BLEU-4 19.62 at epoch 7 (early-stopped); Colab T4.
- Splits: train 6000, val 1000, test 1000.
Known limitation
Attention is effectively 7x7: ResNet-50 at 224px is natively 7x7 and the encoder upsamples to 14x14, so heatmaps are coarse (~32px blocks). Captions are grounded; the spots are region-level, not pixel-level.
Use
huggingface_hub.hf_hub_download("Bukunmi2108/capit-sat", "capit-sat.pt") + vocab.json, then
capit.serving.load_artifact(...).
- Task
- Image-to-Text
- Library
- pytorch
- License
- mit
- Languages
- en
- Created
- Jun 10, 2026
- Updated
- Jun 10, 2026