Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- arampacha/rsicd
|
| 5 |
+
metrics:
|
| 6 |
+
- bleu
|
| 7 |
+
library_name: transformers
|
| 8 |
+
pipeline_tag: image-to-text
|
| 9 |
---
|
| 10 |
+
|
| 11 |
+
# BLIP for RSICD image captioning:
|
| 12 |
+
|
| 13 |
+
- [`blip-image-captioning-base`](https://huggingface.co/Salesforce/blip-image-captioning-base/) model has been finetuned on the [`rsicd`](https://huggingface.co/datasets/arampacha/rsicd) dataset. Training parameters used are as follows:
|
| 14 |
+
- **learning_rate** = 5e-7
|
| 15 |
+
- **optimizer** = AdamW
|
| 16 |
+
- **scheduler** = ReduceLROnPlateau
|
| 17 |
+
- **epochs** = 5
|
| 18 |
+
- More details (demo, testing, evaluation, metrics) available at [`github repo`](https://github.com/gurveervirk/TRINIT-Prometheans-ML03?tab=readme-ov-file)
|