ML Model Accuracy/Results

Handwritten Alphanumeric model

Iteration 1:

Model details
Dataset
Accuracy achieved
Model trained and inference on
Existing dataset
99.80%
Model inference on
NIST dataset
62.70%
Average across all datasets
-
81.25%
Reasoning - As model is trained on the existing dataset, doesn't perform good on NIST dataset

Iteration 2:

Model details
Dataset
Accuracy achieved
Model trained on
Existing dataset + NIST misclassifications
83.30%
Reasoning - After training the model on NIST misclassifications, improvement in accuracy

Iteration 3:

Model details
Dataset
Accuracy achieved
Model trained on
Existing dataset + NIST misclassifications
93.90%
Reasoning - After training the model on NIST misclassifications as per previous checkpoint, improvement in accuracy

Iteration 4:

Model details
Dataset
Accuracy achieved
Model trained on
Existing dataset + manually collected dataset from sheets (~1K samples)
93.90%
Reasoning - Training the model upon previous checkpoint and adding manually collected data, improvement in accuracy

Handwritten Digits model

Iteration 1:

Model details
Dataset
Accuracy achieved
Model trained and inference on
Existing dataset
99.90%
Model inference on
NIST dataset
60.00%
Model inference on
Obtained production data (~50 samples)
97.70%
Average across all datasets
-
85.80%
Reasoning - As model is trained on the existing dataset, doesn't perform good on NIST dataset

Iteration 2:

Model details
Dataset
Accuracy achieved
Model trained on
Existing dataset + NIST misclasifications
96.40%
Reasoning - After training the model on NIST misclassifications, improvement in accuracy

Iteration 3:

Model details
Dataset
Accuracy achieved
Model trained on
Existing dataset + NIST misclasifications + production dataset (~50 samples)
99.70%
Reasoning: After training the model on NIST misclassifications and production dataset, improvement in accuracy

Iteration 4:

Model details
Dataset
Accuracy achieved
Model trained on
Existing dataset + NIST misclasifications + manually collected dataset from sheets (~8.6k samples)
98.30%
Reasoning: As averaging upon a large production dataset, the accuracy slightly dips as compared to iteration 3

Sample dataset images

Existing dataset

Handwritten alphanumeric
​
0a56f9eb1545428f8d9aea0665b7c460
0b0de241b7e64e918fada2f896efdf24
00b6ef24818b4323b4ada81ad433af67
0ac132e3693c47ccb2fe40fb465ba060
0aeaa24048ff4c5d8312d37671c6e08e
0b7f1f188faf453b83704a069791f28f
0bd15fe615db458781bce7dc32a926b5
0adb585eae814b62a56695650cac0666
0b0ffa9c3dcf497c8da259d391f5f159
0b2c4c540a7247629e1dd7a9abd59962
​
Handwritten digits
​
0b1a2b29b0904989a3f6df3a13b93d89
0abb9b222a1142bdaabd30b6742ca8b2
0afd4ac792f14b8185d54cb9d0937b3e
0ac2ad40-7af8-44b4-9829-284a2b33416c_printed
00af4fd32ddc42efbfb1222725599ecf
0aa35cc5-4472-455d-a575-7167c15df849_generated
0a8986b29e804002a3136ddf110f3638
0a1d909f2f2847b795ac40cefa452c3e
0__0040c46a-eae7-4219-bffb-7dd418ad9ffb_up_govt
00c9de07f7b640d083680a5c21661a8b
​

NIST dataset

Handwritten alphanumeric
​
hsf_0_00017
hsf_0_00043
hsf_0_00017
hsf_0_00020
hsf_0_00010
hsf_0_00017
hsf_0_00020
hsf_0_00008
hsf_0_00025
hsf_0_00010
​
Handwritten digits
​
hsf_0_00015
hsf_0_00010
hsf_0_00017
hsf_0_00009
hsf_0_00117
hsf_0_00009
hsf_0_00012
hsf_0_00019
hsf_0_00009
hsf_0_00013
​

Manually collected

Handwritten alphanumeric
​
img014-032
img015-044
img016-045
img023-039
img026-001
img029-045
img013-040
img032-053
img034-055
img011-033
​
Handwritten digits
​
33665
33679
33758
33745
33650
33753
31372
33362
31075
31065
​

Some unhandled misclassifications

​
18103
28926
607
30658
402
32358
24141
30838
24158
32225
​
Reasoning: Generally occurs if the digits are written in corners of the cell