Commit
·
34bd653
1
Parent(s):
e91a99a
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Fine-tuning Details
|
| 2 |
+
"facebook/dino-vitb16" # pre-trained model from which to fine-tune
|
| 3 |
+
|
| 4 |
+
"Graphcore/vit-base-ipu" # config specific to the IPU (Used POD4)
|
| 5 |
+
|
| 6 |
+
Using: [https://github.com/graphcore/Gradient-HuggingFace/tree/main/image-classification](https://github.com/graphcore/Gradient-HuggingFace/commit/826b72cba150be52e7420a3440a31e3096b73c78)
|
| 7 |
+
|
| 8 |
+
Poplar SDK: v3.2.1
|
| 9 |
+
|
| 10 |
+
Dataset:
|
| 11 |
+
|
| 12 |
+
load a custom dataset from local/remote files or folders using the ImageFolder feature
|
| 13 |
+
option 1: local/remote files (supporting the following formats: tar, gzip, zip, xz, rar, zstd)
|
| 14 |
+
url = "https://madm.dfki.de/files/sentinel/EuroSAT.zip"
|
| 15 |
+
files = list(Path(dataset_dir).rglob("EuroSAT.zip"))
|