Image Classification
FBAGSTM commited on
Commit
c0c7289
·
verified ·
1 Parent(s): 7efb952

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -4
README.md CHANGED
@@ -66,7 +66,10 @@ For an image resolution of NxM and P classes
66
 
67
  ## Metrics
68
 
69
- Measures are done with default STM32Cube.AI configuration with enabled input / output allocated option.
 
 
 
70
 
71
 
72
  ### Reference **NPU** memory footprint on food-101 and ImageNet dataset (see Accuracy for details on dataset)
@@ -166,7 +169,7 @@ Dataset details: [link](http://download.tensorflow.org/example_images/flower_pho
166
  ### Accuracy with Plant-village dataset
167
 
168
 
169
- Dataset details: [link](https://data.mendeley.com/datasets/tywbtsjrjv/1) , License [CC0 1.0](https://creativecommons.org/publicdomain/zero/1.0/), Quotation[[2]](#2) , Number of classes: 39, Number of images: 61 486
170
 
171
  | Model | Format | Resolution | Top 1 Accuracy |
172
  |-------|--------|------------|----------------|
@@ -187,7 +190,7 @@ Dataset details: [link](https://data.mendeley.com/datasets/tywbtsjrjv/1) , Licen
187
  ### Accuracy with Food-101 dataset
188
 
189
 
190
- Dataset details: [link](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/) , License [-](), Quotation[[3]](#3) , Number of classes: 101 , Number of images: 101 000
191
 
192
  | Model | Format | Resolution | Top 1 Accuracy |
193
  |-------|--------|------------|----------------|
@@ -209,7 +212,7 @@ Dataset details: [link](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-1
209
 
210
  ### Accuracy with ImageNet dataset
211
 
212
- Dataset details: [link](https://www.image-net.org), License: BSD-3-Clause, Quotation[[4]](#4)
213
  Number of classes: 1000.
214
  To perform the quantization, we calibrated the activations with a random subset of the training set.
215
  For the sake of simplicity, the accuracy reported here was estimated on the 50000 labelled images of the validation set.
 
66
 
67
  ## Metrics
68
 
69
+ - Measures are done with default STM32Cube.AI configuration with enabled input / output allocated option.
70
+ - `tfs` stands for "training from scratch", meaning that the model weights were randomly initialized before training.
71
+ - `tl` stands for "transfer learning", meaning that the model backbone weights were initialized from a pre-trained model, then only the last layer was unfrozen during the training.
72
+ - `fft` stands for "full fine-tuning", meaning that the full model weights were initialized from a transfer learning pre-trained model, and all the layers were unfrozen during the training.
73
 
74
 
75
  ### Reference **NPU** memory footprint on food-101 and ImageNet dataset (see Accuracy for details on dataset)
 
169
  ### Accuracy with Plant-village dataset
170
 
171
 
172
+ Dataset details: [link](https://data.mendeley.com/datasets/tywbtsjrjv/1), License [CC0 1.0](https://creativecommons.org/publicdomain/zero/1.0/), Quotation[[2]](#2) , Number of classes: 39, Number of images: 61 486
173
 
174
  | Model | Format | Resolution | Top 1 Accuracy |
175
  |-------|--------|------------|----------------|
 
190
  ### Accuracy with Food-101 dataset
191
 
192
 
193
+ Dataset details: [link](https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/), Number of classes: 101 , Number of images: 101 000
194
 
195
  | Model | Format | Resolution | Top 1 Accuracy |
196
  |-------|--------|------------|----------------|
 
212
 
213
  ### Accuracy with ImageNet dataset
214
 
215
+ Dataset details: [link](https://www.image-net.org), Quotation[[4]](#4)
216
  Number of classes: 1000.
217
  To perform the quantization, we calibrated the activations with a random subset of the training set.
218
  For the sake of simplicity, the accuracy reported here was estimated on the 50000 labelled images of the validation set.