Add 7 samples with OCR metrics

Model: hoololi/trocr-base-handwritten-calctrainer
Accuracy: 85.7%
Avg OCR time: 0.990s/image
Operations: +: 7
Hardware: ZeroGPU-NVIDIA H200 MIG 3g.71gb

README.md

@@ -65,10 +65,10 @@ dataset_info:
     dtype: float64
   splits:
   - name: train
-    num_bytes: 4372267.125
-    num_examples: 1407
-  download_size: 4218576
-  dataset_size: 4372267.125
+    num_bytes: 4394274.25
+    num_examples: 1414
+  download_size: 4239274
+  dataset_size: 4394274.25
 ---
 
 # CalcTrainer Dataset 🧮

data/train-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d9a01400ddb0d2c1922107a43f0ab57ab9aeaf9a085ee75e31fea76b3ded83a
-size 4218576
+oid sha256:b1d5f89fc9e28a2285be373d17ca9ed41dd8d33a2163b2b45cae07aa713c7e13
+size 4239274