Modified validation and training for linktransformer model

.gitattributes

@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+.git/lfs/objects/8a/f5/8af50c97be5450dea1d079b45919e05e992d4d72babaa4c09a7d64793982cd3e filter=lfs diff=lfs merge=lfs -text

LT_training_config.json

@@ -1,13 +1,13 @@
 {
   "model_save_dir": "models",
   "model_save_name": "linkage_kn_en_familyname",
-  "opt_model_description": "This model was trained on a dataset consisting of 8138064 people and their family id \n                             It was trained for 10 epochs using other defaults that can be found in the repo's LinkTransformer config file - LT_training_config.json \n  ",
+  "opt_model_description": "This model was trained on a dataset consisting of 12105132 people and their family id. 50% of the names are alo transliterated. \n                             It was trained for 6 epochs using other defaults that can be found in the repo's LinkTransformer config file - LT_training_config.json \n  ",
   "opt_model_lang": [
     "kn",
     "en"
   ],
   "train_batch_size": 64,
-  "num_epochs": 10,
+  "num_epochs": 6,
   "warm_up_perc": 1,
   "learning_rate": 2e-06,
   "val_perc": 0.01,

README.md

@@ -24,8 +24,8 @@ This model has been fine-tuned on the model : bert-base-multilingual-cased. It i
 - en.
 
 
-This model was trained on a dataset consisting of 8138064 people and their family id 
-                             It was trained for 10 epochs using other defaults that can be found in the repo's LinkTransformer config file - LT_training_config.json 
+This model was trained on a dataset consisting of 12105132 people and their family id. 50% of the names are alo transliterated. 
+                             It was trained for 6 epochs using other defaults that can be found in the repo's LinkTransformer config file - LT_training_config.json 
   
 
 ## Usage (LinkTransformer)
@@ -99,7 +99,7 @@ The model was trained with the parameters:
 
 **DataLoader**:
 
-`torch.utils.data.dataloader.DataLoader` of length 113924 with parameters:
+`torch.utils.data.dataloader.DataLoader` of length 186000 with parameters:
 ```
 {'batch_size': 64, 'sampler': 'torch.utils.data.dataloader._InfiniteConstantSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
 ```
@@ -111,8 +111,8 @@ The model was trained with the parameters:
 Parameters of the fit()-Method:
 ```
 {
-    "epochs": 10,
-    "evaluation_steps": 11393,
+    "epochs": 6,
+    "evaluation_steps": 18600,
     "evaluator": "sentence_transformers.evaluation.SequentialEvaluator.SequentialEvaluator",
     "max_grad_norm": 1,
     "optimizer_class": "<class 'torch.optim.adamw.AdamW'>",
@@ -121,7 +121,7 @@ Parameters of the fit()-Method:
     },
     "scheduler": "WarmupLinear",
     "steps_per_epoch": null,
-    "warmup_steps": 1139240,
+    "warmup_steps": 1116000,
     "weight_decay": 0.01
 }
 ```

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8af50c97be5450dea1d079b45919e05e992d4d72babaa4c09a7d64793982cd3e
+oid sha256:b729ce841823b35b8245111c07d098e2efaa1e7a93d2e5fe85d5de2bd7806181
 size 711477737