Model save

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: vikp/line_detector_early
+base_model: vikp/line_detector_math
 tags:
 - generated_from_trainer
 model-index:
@@ -12,7 +12,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # line_detector_math
 
-This model is a fine-tuned version of [vikp/line_detector_early](https://huggingface.co/vikp/line_detector_early) on an unknown dataset.
+This model is a fine-tuned version of [vikp/line_detector_math](https://huggingface.co/vikp/line_detector_math) on an unknown dataset.
 
 ## Model description
 
@@ -32,16 +32,16 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 6e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- train_batch_size: 6
+- eval_batch_size: 6
 - seed: 42
 - distributed_type: multi-GPU
 - gradient_accumulation_steps: 8
-- total_train_batch_size: 64
+- total_train_batch_size: 48
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 6
+- num_epochs: 2
 - mixed_precision_training: Native AMP
 
 ### Training results

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "vikp/line_detector_2",
+  "_name_or_path": "vikp/line_detector_math",
   "architectures": [
     "SegformerForRegressionMask"
   ],

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a283299b10560b2faa9836f0de64b597670d8833baafe2be1e025be6b8fcd169
+oid sha256:0e160d03f0e151677db701ac6dcc16599ffe6c0ae1260c82ac6c3087d06b60f9
 size 119881240

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d160591832db656b9e4ca3b3719e0a98c4f3c99938c8d4df44e72a169ad7f01
+oid sha256:3c52870d7d51d23467c93e4a841ad51de11177811ecdc334d940ed903fab8838
 size 4728