DM-Diaz commited on
Commit
d7105b8
·
verified ·
1 Parent(s): 89cf408

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -3
README.md CHANGED
@@ -195,9 +195,9 @@ Because the intercept was added as a final column in the feature matrix, the **f
195
 
196
  ## Internal Model Labels
197
 
198
- The released `.npy` files retain the internal model identifiers used during the original analysis. Public repository names use the terminology reported in the paper.
199
 
200
- | Public model name | Internal identifier |
201
  | --- | --- |
202
  | Baseline | `resnet18-Baseline` |
203
  | Fovea-Gaze | `resnet18-FoveaGaze` |
@@ -207,9 +207,11 @@ The released `.npy` files retain the internal model identifiers used during the
207
  | ImageNet-100 | `resnet18-simclr-imgnet100` |
208
  | ImageNet-1K | `resnet18-simclr-imgnet1k` |
209
 
 
 
210
  The internal identifier `resnet18-pretrained-simclr` corresponds to the STL-10 SimCLR ResNet-18 checkpoint obtained from the external [Spijkervet/SimCLR](https://github.com/Spijkervet/SimCLR) release.
211
 
212
- These internal identifiers are preserved in fields such as `model`, `model1`, `model2`, and the keys of the variance-partitioning dictionaries. They reflect the naming convention used during model development and do not indicate additional model conditions.
213
 
214
  The `features_file_list`, `features_file_list1`, and `features_file_list2` fields also preserve the original analysis paths for provenance. These paths refer to the computing environment used for the original study and are not expected to resolve outside that environment.
215
 
 
195
 
196
  ## Internal Model Labels
197
 
198
+ The released `.npy` files retain the model identifiers used by the original NSD feature-extraction and encoding-model analysis pipeline. These identifiers are analysis-time labels and should not be interpreted as model names stored within the underlying SimCLR checkpoints.
199
 
200
+ | Public model name | Encoding-analysis identifier |
201
  | --- | --- |
202
  | Baseline | `resnet18-Baseline` |
203
  | Fovea-Gaze | `resnet18-FoveaGaze` |
 
207
  | ImageNet-100 | `resnet18-simclr-imgnet100` |
208
  | ImageNet-1K | `resnet18-simclr-imgnet1k` |
209
 
210
+ The ImageNet-100 and ImageNet-1K PyTorch Lightning checkpoints themselves store model parameters under `backbone.*` and `projection_head.*` namespaces and do not contain the dataset-specific analysis identifiers shown above. The STL-10 checkpoint obtained from Spijkervet/SimCLR uses its original `encoder.*` and `projector.*` parameter naming, which was remapped during feature extraction.
211
+
212
  The internal identifier `resnet18-pretrained-simclr` corresponds to the STL-10 SimCLR ResNet-18 checkpoint obtained from the external [Spijkervet/SimCLR](https://github.com/Spijkervet/SimCLR) release.
213
 
214
+ These analysis identifiers are preserved in the `model` field of the released encoding-model fits and, where applicable, in fields such as `model1`, `model2`, and variance-partitioning dictionary keys. They reflect the naming conventions used during model development and do not indicate additional model conditions.
215
 
216
  The `features_file_list`, `features_file_list1`, and `features_file_list2` fields also preserve the original analysis paths for provenance. These paths refer to the computing environment used for the original study and are not expected to resolve outside that environment.
217