rename variants according to changes in paper draft

README.md

@@ -160,8 +160,8 @@ This `Sceptr` object will then have the methods: `calc_pdist_vector`, `calc_cdis
 |`sceptr.variant.tiny`|smaller variant with model dimensionality 16|
 |`sceptr.variant.blosum`|variant using BLOSUM62 embeddings instead of one-hot|
 |`sceptr.variant.average_pooling`|variant using the average-pooling method to generate the TCR representation vector|
-|`sceptr.variant.unpaired`|variant trained on the Tanno et al. dataset with randomised alpha/beta pairing|
-|`sceptr.variant.olga`|variant trained using synthetic TCR sequences generated by OLGA|
+|`sceptr.variant.shuffled_data`|variant trained on the Tanno et al. dataset with randomised alpha/beta pairing|
+|`sceptr.variant.synthetic_data`|variant trained using synthetic TCR sequences generated by OLGA|
 |`sceptr.variant.dropout_noise_only`|variant trained without residue/chain dropping during autocontrastive learning|
 |`sceptr.variant.finetuned`|variant fine-tuned using supervised contrastive learning for six pMHCs with peptides GILGFVFTL, NLVPMVATV, SPRWYFYYL, TFEYVSQPFLMDLE, TTDPSFLGRY and YLQPRTFLL (from [VDJdb](https://vdjdb.cdr3.net/))|
 

src/sceptr/_model_saves/SCEPTR_unpaired/config.json → src/sceptr/_model_saves/SCEPTR_shuffled_data/config.json

@@ -4,7 +4,7 @@
         "initargs": {}
     },
     "model": {
-        "name": "SCEPTR (unpaired)",
+        "name": "SCEPTR (shuffled data)",
         "path_to_pretrained_state_dict": null,
         "token_embedder": {
             "class": "CdrSimpleEmbedder",

src/sceptr/_model_saves/SCEPTR_OLGA/config.json → src/sceptr/_model_saves/SCEPTR_synthetic_data/config.json

@@ -4,7 +4,7 @@
         "initargs": {}
     },
     "model": {
-        "name": "SCEPTR (OLGA)",
+        "name": "SCEPTR (synthetic data)",
         "path_to_pretrained_state_dict": null,
         "token_embedder": {
             "class": "CdrSimpleEmbedder",

src/sceptr/variant.py

@@ -45,12 +45,12 @@ def average_pooling():
     return load_variant("SCEPTR_average_pooling")
 
 
-def unpaired():
-    return load_variant("SCEPTR_unpaired")
+def shuffled_data():
+    return load_variant("SCEPTR_shuffled_data")
 
 
-def olga():
-    return load_variant("SCEPTR_OLGA")
+def synthetic_data():
+    return load_variant("SCEPTR_synthetic_data")
 
 
 def dropout_noise_only():

tests/test_variants.py

@@ -25,8 +25,8 @@ def dummy_data():
         variant.tiny(),
         variant.blosum(),
         variant.average_pooling(),
-        variant.unpaired(),
-        variant.olga(),
+        variant.shuffled_data(),
+        variant.synthetic_data(),
         variant.dropout_noise_only(),
         variant.finetuned(),
         variant.a_sceptr(),