Upload folder using huggingface_hub
Browse files- README.md +396 -0
- config.json +360 -0
- local_compartment_classifier_bd_boxes.skops +0 -0
- train.py +359 -0
README.md
ADDED
|
@@ -0,0 +1,396 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: sklearn
|
| 4 |
+
tags:
|
| 5 |
+
- sklearn
|
| 6 |
+
- skops
|
| 7 |
+
- tabular-classification
|
| 8 |
+
model_format: skops
|
| 9 |
+
model_file: local_compartment_classifier_bd_boxes.skops
|
| 10 |
+
widget:
|
| 11 |
+
- structuredData:
|
| 12 |
+
area_nm2:
|
| 13 |
+
- 693824.0
|
| 14 |
+
- 4852608.0
|
| 15 |
+
- 17088896.0
|
| 16 |
+
area_nm2_neighbor_mean:
|
| 17 |
+
- 10181485.714285716
|
| 18 |
+
- 9884429.714285716
|
| 19 |
+
- 9010409.142857144
|
| 20 |
+
area_nm2_neighbor_std:
|
| 21 |
+
- 8312409.263207569
|
| 22 |
+
- 8587259.418816902
|
| 23 |
+
- 8418630.640116522
|
| 24 |
+
max_dt_nm:
|
| 25 |
+
- 69.0
|
| 26 |
+
- 543.0
|
| 27 |
+
- 1287.0
|
| 28 |
+
max_dt_nm_neighbor_mean:
|
| 29 |
+
- 664.7142857142857
|
| 30 |
+
- 630.8571428571429
|
| 31 |
+
- 577.7142857142857
|
| 32 |
+
max_dt_nm_neighbor_std:
|
| 33 |
+
- 479.64240342658945
|
| 34 |
+
- 504.9563358340017
|
| 35 |
+
- 468.41868657651344
|
| 36 |
+
mean_dt_nm:
|
| 37 |
+
- 24.4375
|
| 38 |
+
- 156.5
|
| 39 |
+
- 416.0
|
| 40 |
+
mean_dt_nm_neighbor_mean:
|
| 41 |
+
- 198.62946428571428
|
| 42 |
+
- 189.19642857142856
|
| 43 |
+
- 170.66071428571428
|
| 44 |
+
mean_dt_nm_neighbor_std:
|
| 45 |
+
- 150.614304054458
|
| 46 |
+
- 157.4368957825056
|
| 47 |
+
- 143.32375093543624
|
| 48 |
+
pca_ratio_01:
|
| 49 |
+
- 1.3849340770961909
|
| 50 |
+
- 1.181656878273399
|
| 51 |
+
- 1.128046800200765
|
| 52 |
+
pca_ratio_01_neighbor_mean:
|
| 53 |
+
- 1.8575624906424115
|
| 54 |
+
- 1.8760422359899387
|
| 55 |
+
- 1.880915879451087
|
| 56 |
+
pca_ratio_01_neighbor_std:
|
| 57 |
+
- 0.641580757345606
|
| 58 |
+
- 0.6228187048854344
|
| 59 |
+
- 0.6165585104590592
|
| 60 |
+
pca_unwrapped_0:
|
| 61 |
+
- -0.0046539306640625
|
| 62 |
+
- -0.497314453125
|
| 63 |
+
- -0.258544921875
|
| 64 |
+
pca_unwrapped_0_neighbor_mean:
|
| 65 |
+
- 0.039224624633789
|
| 66 |
+
- 0.0840119448575106
|
| 67 |
+
- 0.0623056238347833
|
| 68 |
+
pca_unwrapped_0_neighbor_std:
|
| 69 |
+
- 0.3114910605258688
|
| 70 |
+
- 0.2573427692683507
|
| 71 |
+
- 0.296254177168357
|
| 72 |
+
pca_unwrapped_1:
|
| 73 |
+
- 0.7392578125
|
| 74 |
+
- -0.11553955078125
|
| 75 |
+
- 0.2169189453125
|
| 76 |
+
pca_unwrapped_1_neighbor_mean:
|
| 77 |
+
- 0.0941687497225674
|
| 78 |
+
- 0.1718776009299538
|
| 79 |
+
- 0.1416541012850674
|
| 80 |
+
pca_unwrapped_1_neighbor_std:
|
| 81 |
+
- 0.3179467337379631
|
| 82 |
+
- 0.3628551035117971
|
| 83 |
+
- 0.372447324946889
|
| 84 |
+
pca_unwrapped_2:
|
| 85 |
+
- -0.673828125
|
| 86 |
+
- -0.85986328125
|
| 87 |
+
- 0.94140625
|
| 88 |
+
pca_unwrapped_2_neighbor_mean:
|
| 89 |
+
- 0.2258744673295454
|
| 90 |
+
- 0.2427867542613636
|
| 91 |
+
- 0.0790349786931818
|
| 92 |
+
pca_unwrapped_2_neighbor_std:
|
| 93 |
+
- 0.9134250264562896
|
| 94 |
+
- 0.8928014788058292
|
| 95 |
+
- 0.9167197839332804
|
| 96 |
+
pca_unwrapped_3:
|
| 97 |
+
- -0.0302886962890625
|
| 98 |
+
- -0.86572265625
|
| 99 |
+
- 0.57177734375
|
| 100 |
+
pca_unwrapped_3_neighbor_mean:
|
| 101 |
+
- -0.2933238636363636
|
| 102 |
+
- -0.2173753218217329
|
| 103 |
+
- -0.3480571400035511
|
| 104 |
+
pca_unwrapped_3_neighbor_std:
|
| 105 |
+
- 0.6203425764161097
|
| 106 |
+
- 0.5938304683645145
|
| 107 |
+
- 0.5600074530240728
|
| 108 |
+
pca_unwrapped_4:
|
| 109 |
+
- 0.67333984375
|
| 110 |
+
- -0.0005474090576171
|
| 111 |
+
- 0.81982421875
|
| 112 |
+
pca_unwrapped_4_neighbor_mean:
|
| 113 |
+
- 0.2915762121027166
|
| 114 |
+
- 0.3528386896306818
|
| 115 |
+
- 0.2782594507390802
|
| 116 |
+
pca_unwrapped_4_neighbor_std:
|
| 117 |
+
- 0.6415192812587974
|
| 118 |
+
- 0.6430080201673403
|
| 119 |
+
- 0.6308895861182334
|
| 120 |
+
pca_unwrapped_5:
|
| 121 |
+
- 0.73876953125
|
| 122 |
+
- 0.50048828125
|
| 123 |
+
- -0.03192138671875
|
| 124 |
+
pca_unwrapped_5_neighbor_mean:
|
| 125 |
+
- 0.2028697620738636
|
| 126 |
+
- 0.2245316938920454
|
| 127 |
+
- 0.2729325727982954
|
| 128 |
+
pca_unwrapped_5_neighbor_std:
|
| 129 |
+
- 0.265173781606759
|
| 130 |
+
- 0.2994363858938455
|
| 131 |
+
- 0.2968562365279343
|
| 132 |
+
pca_unwrapped_6:
|
| 133 |
+
- 0.99951171875
|
| 134 |
+
- 0.05828857421875
|
| 135 |
+
- -0.77880859375
|
| 136 |
+
pca_unwrapped_6_neighbor_mean:
|
| 137 |
+
- -0.2386505820534446
|
| 138 |
+
- -0.1530848416415128
|
| 139 |
+
- -0.0769850990988991
|
| 140 |
+
pca_unwrapped_6_neighbor_std:
|
| 141 |
+
- 0.6776577717043619
|
| 142 |
+
- 0.7717860533115238
|
| 143 |
+
- 0.7447135522384378
|
| 144 |
+
pca_unwrapped_7:
|
| 145 |
+
- 0.023834228515625
|
| 146 |
+
- -0.9931640625
|
| 147 |
+
- 0.52978515625
|
| 148 |
+
pca_unwrapped_7_neighbor_mean:
|
| 149 |
+
- -0.4803272594105113
|
| 150 |
+
- -0.3878728693181818
|
| 151 |
+
- -0.5263227982954546
|
| 152 |
+
pca_unwrapped_7_neighbor_std:
|
| 153 |
+
- 0.4799926318285017
|
| 154 |
+
- 0.4691567465869561
|
| 155 |
+
- 0.3891669942534205
|
| 156 |
+
pca_unwrapped_8:
|
| 157 |
+
- 0.0192413330078125
|
| 158 |
+
- 0.0997314453125
|
| 159 |
+
- -0.3359375
|
| 160 |
+
pca_unwrapped_8_neighbor_mean:
|
| 161 |
+
- -0.0384375832297585
|
| 162 |
+
- -0.0457548661665482
|
| 163 |
+
- -0.0061485984108664
|
| 164 |
+
pca_unwrapped_8_neighbor_std:
|
| 165 |
+
- 0.3037878488292577
|
| 166 |
+
- 0.3010843368506175
|
| 167 |
+
- 0.2874409267860334
|
| 168 |
+
pca_val_unwrapped_0:
|
| 169 |
+
- 15657.09765625
|
| 170 |
+
- 40668.40625
|
| 171 |
+
- 66863.0
|
| 172 |
+
pca_val_unwrapped_0_neighbor_mean:
|
| 173 |
+
- 69378.52059659091
|
| 174 |
+
- 67104.76526988637
|
| 175 |
+
- 64723.43856534091
|
| 176 |
+
pca_val_unwrapped_0_neighbor_std:
|
| 177 |
+
- 20242.245019019712
|
| 178 |
+
- 24702.906417865197
|
| 179 |
+
- 25959.16138296664
|
| 180 |
+
pca_val_unwrapped_1:
|
| 181 |
+
- 11305.3017578125
|
| 182 |
+
- 34416.42578125
|
| 183 |
+
- 59273.25
|
| 184 |
+
pca_val_unwrapped_1_neighbor_mean:
|
| 185 |
+
- 41190.40261008523
|
| 186 |
+
- 39089.39133522727
|
| 187 |
+
- 36829.68004261364
|
| 188 |
+
pca_val_unwrapped_1_neighbor_std:
|
| 189 |
+
- 16625.870141811894
|
| 190 |
+
- 18875.56976212627
|
| 191 |
+
- 17666.778281657556
|
| 192 |
+
pca_val_unwrapped_2:
|
| 193 |
+
- 1270.4095458984375
|
| 194 |
+
- 13551.6748046875
|
| 195 |
+
- 47764.625
|
| 196 |
+
pca_val_unwrapped_2_neighbor_mean:
|
| 197 |
+
- 28717.50048828125
|
| 198 |
+
- 27601.021828391335
|
| 199 |
+
- 24490.75362881747
|
| 200 |
+
pca_val_unwrapped_2_neighbor_std:
|
| 201 |
+
- 14988.204981576571
|
| 202 |
+
- 16601.48080038032
|
| 203 |
+
- 15622.078784778376
|
| 204 |
+
post_synapse_count:
|
| 205 |
+
- 0.0
|
| 206 |
+
- 0.0
|
| 207 |
+
- 0.0
|
| 208 |
+
post_synapse_count_neighbor_mean:
|
| 209 |
+
- 0.0
|
| 210 |
+
- 0.0
|
| 211 |
+
- 0.0
|
| 212 |
+
post_synapse_count_neighbor_std:
|
| 213 |
+
- 0.0
|
| 214 |
+
- 0.0
|
| 215 |
+
- 0.0
|
| 216 |
+
pre_synapse_count:
|
| 217 |
+
- 0.0
|
| 218 |
+
- 0.0
|
| 219 |
+
- 0.0
|
| 220 |
+
pre_synapse_count_neighbor_mean:
|
| 221 |
+
- 0.0
|
| 222 |
+
- 0.0
|
| 223 |
+
- 0.0
|
| 224 |
+
pre_synapse_count_neighbor_std:
|
| 225 |
+
- 0.0
|
| 226 |
+
- 0.0
|
| 227 |
+
- 0.0
|
| 228 |
+
size_nm3:
|
| 229 |
+
- 12771840.0
|
| 230 |
+
- 697943040.0
|
| 231 |
+
- 7550330880.0
|
| 232 |
+
size_nm3_neighbor_mean:
|
| 233 |
+
- 3233702034.285714
|
| 234 |
+
- 3184761234.285714
|
| 235 |
+
- 2695304960.0
|
| 236 |
+
size_nm3_neighbor_std:
|
| 237 |
+
- 3650678969.7909584
|
| 238 |
+
- 3691650923.5639486
|
| 239 |
+
- 3518520747.0511127
|
| 240 |
+
---
|
| 241 |
+
|
| 242 |
+
# Model description
|
| 243 |
+
|
| 244 |
+
[More Information Needed]
|
| 245 |
+
|
| 246 |
+
## Intended uses & limitations
|
| 247 |
+
|
| 248 |
+
[More Information Needed]
|
| 249 |
+
|
| 250 |
+
## Training Procedure
|
| 251 |
+
|
| 252 |
+
[More Information Needed]
|
| 253 |
+
|
| 254 |
+
### Hyperparameters
|
| 255 |
+
|
| 256 |
+
<details>
|
| 257 |
+
<summary> Click to expand </summary>
|
| 258 |
+
|
| 259 |
+
| Hyperparameter | Value |
|
| 260 |
+
|------------------------------------|---------------------------------------------------------------------------------------------------------------------------|
|
| 261 |
+
| memory | |
|
| 262 |
+
| steps | [('transformer', QuantileTransformer(output_distribution='normal')), ('lda', LinearDiscriminantAnalysis(n_components=3))] |
|
| 263 |
+
| verbose | False |
|
| 264 |
+
| transformer | QuantileTransformer(output_distribution='normal') |
|
| 265 |
+
| lda | LinearDiscriminantAnalysis(n_components=3) |
|
| 266 |
+
| transformer__copy | True |
|
| 267 |
+
| transformer__ignore_implicit_zeros | False |
|
| 268 |
+
| transformer__n_quantiles | 1000 |
|
| 269 |
+
| transformer__output_distribution | normal |
|
| 270 |
+
| transformer__random_state | |
|
| 271 |
+
| transformer__subsample | 10000 |
|
| 272 |
+
| lda__covariance_estimator | |
|
| 273 |
+
| lda__n_components | 3 |
|
| 274 |
+
| lda__priors | |
|
| 275 |
+
| lda__shrinkage | |
|
| 276 |
+
| lda__solver | svd |
|
| 277 |
+
| lda__store_covariance | False |
|
| 278 |
+
| lda__tol | 0.0001 |
|
| 279 |
+
|
| 280 |
+
</details>
|
| 281 |
+
|
| 282 |
+
### Model Plot
|
| 283 |
+
|
| 284 |
+
<style>#sk-container-id-4 {/* Definition of color scheme common for light and dark mode */--sklearn-color-text: black;--sklearn-color-line: gray;/* Definition of color scheme for unfitted estimators */--sklearn-color-unfitted-level-0: #fff5e6;--sklearn-color-unfitted-level-1: #f6e4d2;--sklearn-color-unfitted-level-2: #ffe0b3;--sklearn-color-unfitted-level-3: chocolate;/* Definition of color scheme for fitted estimators */--sklearn-color-fitted-level-0: #f0f8ff;--sklearn-color-fitted-level-1: #d4ebff;--sklearn-color-fitted-level-2: #b3dbfd;--sklearn-color-fitted-level-3: cornflowerblue;/* Specific color for light theme */--sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));--sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));--sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));--sklearn-color-icon: #696969;@media (prefers-color-scheme: dark) {/* Redefinition of color scheme for dark theme */--sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));--sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));--sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));--sklearn-color-icon: #878787;}
|
| 285 |
+
}#sk-container-id-4 {color: var(--sklearn-color-text);
|
| 286 |
+
}#sk-container-id-4 pre {padding: 0;
|
| 287 |
+
}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;
|
| 288 |
+
}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed var(--sklearn-color-line);margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: var(--sklearn-color-background);
|
| 289 |
+
}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }`but bootstrap.min.css set `[hidden] { display: none !important; }`so we also need the `!important` here to be able to override thedefault hidden behavior on the sphinx rendered scikit-learn.org.See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;
|
| 290 |
+
}#sk-container-id-4 div.sk-text-repr-fallback {display: none;
|
| 291 |
+
}div.sk-parallel-item,
|
| 292 |
+
div.sk-serial,
|
| 293 |
+
div.sk-item {/* draw centered vertical line to link estimators */background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));background-size: 2px 100%;background-repeat: no-repeat;background-position: center center;
|
| 294 |
+
}/* Parallel-specific style estimator block */#sk-container-id-4 div.sk-parallel-item::after {content: "";width: 100%;border-bottom: 2px solid var(--sklearn-color-text-on-default-background);flex-grow: 1;
|
| 295 |
+
}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: var(--sklearn-color-background);position: relative;
|
| 296 |
+
}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;
|
| 297 |
+
}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;
|
| 298 |
+
}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;
|
| 299 |
+
}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;
|
| 300 |
+
}/* Serial-specific style estimator block */#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: var(--sklearn-color-background);padding-right: 1em;padding-left: 1em;
|
| 301 |
+
}/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is
|
| 302 |
+
clickable and can be expanded/collapsed.
|
| 303 |
+
- Pipeline and ColumnTransformer use this feature and define the default style
|
| 304 |
+
- Estimators will overwrite some part of the style using the `sk-estimator` class
|
| 305 |
+
*//* Pipeline and ColumnTransformer style (default) */#sk-container-id-4 div.sk-toggleable {/* Default theme specific background. It is overwritten whether we have aspecific estimator or a Pipeline/ColumnTransformer */background-color: var(--sklearn-color-background);
|
| 306 |
+
}/* Toggleable label */
|
| 307 |
+
#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.5em;box-sizing: border-box;text-align: center;
|
| 308 |
+
}#sk-container-id-4 label.sk-toggleable__label-arrow:before {/* Arrow on the left of the label */content: "▸";float: left;margin-right: 0.25em;color: var(--sklearn-color-icon);
|
| 309 |
+
}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: var(--sklearn-color-text);
|
| 310 |
+
}/* Toggleable content - dropdown */#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;/* unfitted */background-color: var(--sklearn-color-unfitted-level-0);
|
| 311 |
+
}#sk-container-id-4 div.sk-toggleable__content.fitted {/* fitted */background-color: var(--sklearn-color-fitted-level-0);
|
| 312 |
+
}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;border-radius: 0.25em;color: var(--sklearn-color-text);/* unfitted */background-color: var(--sklearn-color-unfitted-level-0);
|
| 313 |
+
}#sk-container-id-4 div.sk-toggleable__content.fitted pre {/* unfitted */background-color: var(--sklearn-color-fitted-level-0);
|
| 314 |
+
}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {/* Expand drop-down */max-height: 200px;max-width: 100%;overflow: auto;
|
| 315 |
+
}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: "▾";
|
| 316 |
+
}/* Pipeline/ColumnTransformer-specific style */#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {color: var(--sklearn-color-text);background-color: var(--sklearn-color-unfitted-level-2);
|
| 317 |
+
}#sk-container-id-4 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: var(--sklearn-color-fitted-level-2);
|
| 318 |
+
}/* Estimator-specific style *//* Colorize estimator box */
|
| 319 |
+
#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {/* unfitted */background-color: var(--sklearn-color-unfitted-level-2);
|
| 320 |
+
}#sk-container-id-4 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {/* fitted */background-color: var(--sklearn-color-fitted-level-2);
|
| 321 |
+
}#sk-container-id-4 div.sk-label label.sk-toggleable__label,
|
| 322 |
+
#sk-container-id-4 div.sk-label label {/* The background is the default theme color */color: var(--sklearn-color-text-on-default-background);
|
| 323 |
+
}/* On hover, darken the color of the background */
|
| 324 |
+
#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {color: var(--sklearn-color-text);background-color: var(--sklearn-color-unfitted-level-2);
|
| 325 |
+
}/* Label box, darken color on hover, fitted */
|
| 326 |
+
#sk-container-id-4 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {color: var(--sklearn-color-text);background-color: var(--sklearn-color-fitted-level-2);
|
| 327 |
+
}/* Estimator label */#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;
|
| 328 |
+
}#sk-container-id-4 div.sk-label-container {text-align: center;
|
| 329 |
+
}/* Estimator-specific */
|
| 330 |
+
#sk-container-id-4 div.sk-estimator {font-family: monospace;border: 1px dotted var(--sklearn-color-border-box);border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;/* unfitted */background-color: var(--sklearn-color-unfitted-level-0);
|
| 331 |
+
}#sk-container-id-4 div.sk-estimator.fitted {/* fitted */background-color: var(--sklearn-color-fitted-level-0);
|
| 332 |
+
}/* on hover */
|
| 333 |
+
#sk-container-id-4 div.sk-estimator:hover {/* unfitted */background-color: var(--sklearn-color-unfitted-level-2);
|
| 334 |
+
}#sk-container-id-4 div.sk-estimator.fitted:hover {/* fitted */background-color: var(--sklearn-color-fitted-level-2);
|
| 335 |
+
}/* Specification for estimator info (e.g. "i" and "?") *//* Common style for "i" and "?" */.sk-estimator-doc-link,
|
| 336 |
+
a:link.sk-estimator-doc-link,
|
| 337 |
+
a:visited.sk-estimator-doc-link {float: right;font-size: smaller;line-height: 1em;font-family: monospace;background-color: var(--sklearn-color-background);border-radius: 1em;height: 1em;width: 1em;text-decoration: none !important;margin-left: 1ex;/* unfitted */border: var(--sklearn-color-unfitted-level-1) 1pt solid;color: var(--sklearn-color-unfitted-level-1);
|
| 338 |
+
}.sk-estimator-doc-link.fitted,
|
| 339 |
+
a:link.sk-estimator-doc-link.fitted,
|
| 340 |
+
a:visited.sk-estimator-doc-link.fitted {/* fitted */border: var(--sklearn-color-fitted-level-1) 1pt solid;color: var(--sklearn-color-fitted-level-1);
|
| 341 |
+
}/* On hover */
|
| 342 |
+
div.sk-estimator:hover .sk-estimator-doc-link:hover,
|
| 343 |
+
.sk-estimator-doc-link:hover,
|
| 344 |
+
div.sk-label-container:hover .sk-estimator-doc-link:hover,
|
| 345 |
+
.sk-estimator-doc-link:hover {/* unfitted */background-color: var(--sklearn-color-unfitted-level-3);color: var(--sklearn-color-background);text-decoration: none;
|
| 346 |
+
}div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,
|
| 347 |
+
.sk-estimator-doc-link.fitted:hover,
|
| 348 |
+
div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,
|
| 349 |
+
.sk-estimator-doc-link.fitted:hover {/* fitted */background-color: var(--sklearn-color-fitted-level-3);color: var(--sklearn-color-background);text-decoration: none;
|
| 350 |
+
}/* Span, style for the box shown on hovering the info icon */
|
| 351 |
+
.sk-estimator-doc-link span {display: none;z-index: 9999;position: relative;font-weight: normal;right: .2ex;padding: .5ex;margin: .5ex;width: min-content;min-width: 20ex;max-width: 50ex;color: var(--sklearn-color-text);box-shadow: 2pt 2pt 4pt #999;/* unfitted */background: var(--sklearn-color-unfitted-level-0);border: .5pt solid var(--sklearn-color-unfitted-level-3);
|
| 352 |
+
}.sk-estimator-doc-link.fitted span {/* fitted */background: var(--sklearn-color-fitted-level-0);border: var(--sklearn-color-fitted-level-3);
|
| 353 |
+
}.sk-estimator-doc-link:hover span {display: block;
|
| 354 |
+
}/* "?"-specific style due to the `<a>` HTML tag */#sk-container-id-4 a.estimator_doc_link {float: right;font-size: 1rem;line-height: 1em;font-family: monospace;background-color: var(--sklearn-color-background);border-radius: 1rem;height: 1rem;width: 1rem;text-decoration: none;/* unfitted */color: var(--sklearn-color-unfitted-level-1);border: var(--sklearn-color-unfitted-level-1) 1pt solid;
|
| 355 |
+
}#sk-container-id-4 a.estimator_doc_link.fitted {/* fitted */border: var(--sklearn-color-fitted-level-1) 1pt solid;color: var(--sklearn-color-fitted-level-1);
|
| 356 |
+
}/* On hover */
|
| 357 |
+
#sk-container-id-4 a.estimator_doc_link:hover {/* unfitted */background-color: var(--sklearn-color-unfitted-level-3);color: var(--sklearn-color-background);text-decoration: none;
|
| 358 |
+
}#sk-container-id-4 a.estimator_doc_link.fitted:hover {/* fitted */background-color: var(--sklearn-color-fitted-level-3);
|
| 359 |
+
}
|
| 360 |
+
</style><div id="sk-container-id-4" class="sk-top-container" style="overflow: auto;"><div class="sk-text-repr-fallback"><pre>Pipeline(steps=[('transformer',QuantileTransformer(output_distribution='normal')),('lda', LinearDiscriminantAnalysis(n_components=3))])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class="sk-container" hidden><div class="sk-item sk-dashed-wrapped"><div class="sk-label-container"><div class="sk-label fitted sk-toggleable"><input class="sk-toggleable__control sk-hidden--visually" id="sk-estimator-id-10" type="checkbox" ><label for="sk-estimator-id-10" class="sk-toggleable__label fitted sk-toggleable__label-arrow fitted"> Pipeline<a class="sk-estimator-doc-link fitted" rel="noreferrer" target="_blank" href="https://scikit-learn.org/1.4/modules/generated/sklearn.pipeline.Pipeline.html">?<span>Documentation for Pipeline</span></a><span class="sk-estimator-doc-link fitted">i<span>Fitted</span></span></label><div class="sk-toggleable__content fitted"><pre>Pipeline(steps=[('transformer',QuantileTransformer(output_distribution='normal')),('lda', LinearDiscriminantAnalysis(n_components=3))])</pre></div> </div></div><div class="sk-serial"><div class="sk-item"><div class="sk-estimator fitted sk-toggleable"><input class="sk-toggleable__control sk-hidden--visually" id="sk-estimator-id-11" type="checkbox" ><label for="sk-estimator-id-11" class="sk-toggleable__label fitted sk-toggleable__label-arrow fitted"> QuantileTransformer<a class="sk-estimator-doc-link fitted" rel="noreferrer" target="_blank" href="https://scikit-learn.org/1.4/modules/generated/sklearn.preprocessing.QuantileTransformer.html">?<span>Documentation for QuantileTransformer</span></a></label><div class="sk-toggleable__content fitted"><pre>QuantileTransformer(output_distribution='normal')</pre></div> </div></div><div class="sk-item"><div class="sk-estimator fitted sk-toggleable"><input class="sk-toggleable__control sk-hidden--visually" id="sk-estimator-id-12" type="checkbox" ><label for="sk-estimator-id-12" class="sk-toggleable__label fitted sk-toggleable__label-arrow fitted"> LinearDiscriminantAnalysis<a class="sk-estimator-doc-link fitted" rel="noreferrer" target="_blank" href="https://scikit-learn.org/1.4/modules/generated/sklearn.discriminant_analysis.LinearDiscriminantAnalysis.html">?<span>Documentation for LinearDiscriminantAnalysis</span></a></label><div class="sk-toggleable__content fitted"><pre>LinearDiscriminantAnalysis(n_components=3)</pre></div> </div></div></div></div></div></div>
|
| 361 |
+
|
| 362 |
+
## Evaluation Results
|
| 363 |
+
|
| 364 |
+
[More Information Needed]
|
| 365 |
+
|
| 366 |
+
# How to Get Started with the Model
|
| 367 |
+
|
| 368 |
+
[More Information Needed]
|
| 369 |
+
|
| 370 |
+
# Model Card Authors
|
| 371 |
+
|
| 372 |
+
This model card is written by following authors:
|
| 373 |
+
|
| 374 |
+
[More Information Needed]
|
| 375 |
+
|
| 376 |
+
# Model Card Contact
|
| 377 |
+
|
| 378 |
+
You can contact the model card authors through following channels:
|
| 379 |
+
[More Information Needed]
|
| 380 |
+
|
| 381 |
+
# Citation
|
| 382 |
+
|
| 383 |
+
Below you can find information related to citation.
|
| 384 |
+
|
| 385 |
+
**BibTeX:**
|
| 386 |
+
```
|
| 387 |
+
[More Information Needed]
|
| 388 |
+
```
|
| 389 |
+
|
| 390 |
+
# model_card_authors
|
| 391 |
+
|
| 392 |
+
bdpedigo
|
| 393 |
+
|
| 394 |
+
# model_description
|
| 395 |
+
|
| 396 |
+
This is a model trained to classify pieces of neuron as axon, dendrite, soma, orglia, based only on their local shape and synapse features.The model is a linear discriminant classifier which was trained on compartment labels generated by Bethanny Danskin for 3 6x6x6 um boxes in the Minnie65 Phase3 dataset.
|
config.json
ADDED
|
@@ -0,0 +1,360 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"sklearn": {
|
| 3 |
+
"columns": [
|
| 4 |
+
"area_nm2",
|
| 5 |
+
"max_dt_nm",
|
| 6 |
+
"mean_dt_nm",
|
| 7 |
+
"size_nm3",
|
| 8 |
+
"pca_unwrapped_0",
|
| 9 |
+
"pca_unwrapped_1",
|
| 10 |
+
"pca_unwrapped_2",
|
| 11 |
+
"pca_unwrapped_3",
|
| 12 |
+
"pca_unwrapped_4",
|
| 13 |
+
"pca_unwrapped_5",
|
| 14 |
+
"pca_unwrapped_6",
|
| 15 |
+
"pca_unwrapped_7",
|
| 16 |
+
"pca_unwrapped_8",
|
| 17 |
+
"pca_val_unwrapped_0",
|
| 18 |
+
"pca_val_unwrapped_1",
|
| 19 |
+
"pca_val_unwrapped_2",
|
| 20 |
+
"pca_ratio_01",
|
| 21 |
+
"pre_synapse_count",
|
| 22 |
+
"post_synapse_count",
|
| 23 |
+
"area_nm2_neighbor_mean",
|
| 24 |
+
"area_nm2_neighbor_std",
|
| 25 |
+
"max_dt_nm_neighbor_mean",
|
| 26 |
+
"max_dt_nm_neighbor_std",
|
| 27 |
+
"mean_dt_nm_neighbor_mean",
|
| 28 |
+
"mean_dt_nm_neighbor_std",
|
| 29 |
+
"size_nm3_neighbor_mean",
|
| 30 |
+
"size_nm3_neighbor_std",
|
| 31 |
+
"pca_unwrapped_0_neighbor_mean",
|
| 32 |
+
"pca_unwrapped_0_neighbor_std",
|
| 33 |
+
"pca_unwrapped_1_neighbor_mean",
|
| 34 |
+
"pca_unwrapped_1_neighbor_std",
|
| 35 |
+
"pca_unwrapped_2_neighbor_mean",
|
| 36 |
+
"pca_unwrapped_2_neighbor_std",
|
| 37 |
+
"pca_unwrapped_3_neighbor_mean",
|
| 38 |
+
"pca_unwrapped_3_neighbor_std",
|
| 39 |
+
"pca_unwrapped_4_neighbor_mean",
|
| 40 |
+
"pca_unwrapped_4_neighbor_std",
|
| 41 |
+
"pca_unwrapped_5_neighbor_mean",
|
| 42 |
+
"pca_unwrapped_5_neighbor_std",
|
| 43 |
+
"pca_unwrapped_6_neighbor_mean",
|
| 44 |
+
"pca_unwrapped_6_neighbor_std",
|
| 45 |
+
"pca_unwrapped_7_neighbor_mean",
|
| 46 |
+
"pca_unwrapped_7_neighbor_std",
|
| 47 |
+
"pca_unwrapped_8_neighbor_mean",
|
| 48 |
+
"pca_unwrapped_8_neighbor_std",
|
| 49 |
+
"pca_val_unwrapped_0_neighbor_mean",
|
| 50 |
+
"pca_val_unwrapped_0_neighbor_std",
|
| 51 |
+
"pca_val_unwrapped_1_neighbor_mean",
|
| 52 |
+
"pca_val_unwrapped_1_neighbor_std",
|
| 53 |
+
"pca_val_unwrapped_2_neighbor_mean",
|
| 54 |
+
"pca_val_unwrapped_2_neighbor_std",
|
| 55 |
+
"pca_ratio_01_neighbor_mean",
|
| 56 |
+
"pca_ratio_01_neighbor_std",
|
| 57 |
+
"pre_synapse_count_neighbor_mean",
|
| 58 |
+
"pre_synapse_count_neighbor_std",
|
| 59 |
+
"post_synapse_count_neighbor_mean",
|
| 60 |
+
"post_synapse_count_neighbor_std"
|
| 61 |
+
],
|
| 62 |
+
"environment": [
|
| 63 |
+
"scikit-learn",
|
| 64 |
+
"caveclient"
|
| 65 |
+
],
|
| 66 |
+
"example_input": {
|
| 67 |
+
"area_nm2": [
|
| 68 |
+
693824.0,
|
| 69 |
+
4852608.0,
|
| 70 |
+
17088896.0
|
| 71 |
+
],
|
| 72 |
+
"area_nm2_neighbor_mean": [
|
| 73 |
+
10181485.714285716,
|
| 74 |
+
9884429.714285716,
|
| 75 |
+
9010409.142857144
|
| 76 |
+
],
|
| 77 |
+
"area_nm2_neighbor_std": [
|
| 78 |
+
8312409.263207569,
|
| 79 |
+
8587259.418816902,
|
| 80 |
+
8418630.640116522
|
| 81 |
+
],
|
| 82 |
+
"max_dt_nm": [
|
| 83 |
+
69.0,
|
| 84 |
+
543.0,
|
| 85 |
+
1287.0
|
| 86 |
+
],
|
| 87 |
+
"max_dt_nm_neighbor_mean": [
|
| 88 |
+
664.7142857142857,
|
| 89 |
+
630.8571428571429,
|
| 90 |
+
577.7142857142857
|
| 91 |
+
],
|
| 92 |
+
"max_dt_nm_neighbor_std": [
|
| 93 |
+
479.64240342658945,
|
| 94 |
+
504.9563358340017,
|
| 95 |
+
468.41868657651344
|
| 96 |
+
],
|
| 97 |
+
"mean_dt_nm": [
|
| 98 |
+
24.4375,
|
| 99 |
+
156.5,
|
| 100 |
+
416.0
|
| 101 |
+
],
|
| 102 |
+
"mean_dt_nm_neighbor_mean": [
|
| 103 |
+
198.62946428571428,
|
| 104 |
+
189.19642857142856,
|
| 105 |
+
170.66071428571428
|
| 106 |
+
],
|
| 107 |
+
"mean_dt_nm_neighbor_std": [
|
| 108 |
+
150.614304054458,
|
| 109 |
+
157.4368957825056,
|
| 110 |
+
143.32375093543624
|
| 111 |
+
],
|
| 112 |
+
"pca_ratio_01": [
|
| 113 |
+
1.3849340770961909,
|
| 114 |
+
1.181656878273399,
|
| 115 |
+
1.128046800200765
|
| 116 |
+
],
|
| 117 |
+
"pca_ratio_01_neighbor_mean": [
|
| 118 |
+
1.8575624906424115,
|
| 119 |
+
1.8760422359899387,
|
| 120 |
+
1.880915879451087
|
| 121 |
+
],
|
| 122 |
+
"pca_ratio_01_neighbor_std": [
|
| 123 |
+
0.641580757345606,
|
| 124 |
+
0.6228187048854344,
|
| 125 |
+
0.6165585104590592
|
| 126 |
+
],
|
| 127 |
+
"pca_unwrapped_0": [
|
| 128 |
+
-0.0046539306640625,
|
| 129 |
+
-0.497314453125,
|
| 130 |
+
-0.258544921875
|
| 131 |
+
],
|
| 132 |
+
"pca_unwrapped_0_neighbor_mean": [
|
| 133 |
+
0.039224624633789,
|
| 134 |
+
0.0840119448575106,
|
| 135 |
+
0.0623056238347833
|
| 136 |
+
],
|
| 137 |
+
"pca_unwrapped_0_neighbor_std": [
|
| 138 |
+
0.3114910605258688,
|
| 139 |
+
0.2573427692683507,
|
| 140 |
+
0.296254177168357
|
| 141 |
+
],
|
| 142 |
+
"pca_unwrapped_1": [
|
| 143 |
+
0.7392578125,
|
| 144 |
+
-0.11553955078125,
|
| 145 |
+
0.2169189453125
|
| 146 |
+
],
|
| 147 |
+
"pca_unwrapped_1_neighbor_mean": [
|
| 148 |
+
0.0941687497225674,
|
| 149 |
+
0.1718776009299538,
|
| 150 |
+
0.1416541012850674
|
| 151 |
+
],
|
| 152 |
+
"pca_unwrapped_1_neighbor_std": [
|
| 153 |
+
0.3179467337379631,
|
| 154 |
+
0.3628551035117971,
|
| 155 |
+
0.372447324946889
|
| 156 |
+
],
|
| 157 |
+
"pca_unwrapped_2": [
|
| 158 |
+
-0.673828125,
|
| 159 |
+
-0.85986328125,
|
| 160 |
+
0.94140625
|
| 161 |
+
],
|
| 162 |
+
"pca_unwrapped_2_neighbor_mean": [
|
| 163 |
+
0.2258744673295454,
|
| 164 |
+
0.2427867542613636,
|
| 165 |
+
0.0790349786931818
|
| 166 |
+
],
|
| 167 |
+
"pca_unwrapped_2_neighbor_std": [
|
| 168 |
+
0.9134250264562896,
|
| 169 |
+
0.8928014788058292,
|
| 170 |
+
0.9167197839332804
|
| 171 |
+
],
|
| 172 |
+
"pca_unwrapped_3": [
|
| 173 |
+
-0.0302886962890625,
|
| 174 |
+
-0.86572265625,
|
| 175 |
+
0.57177734375
|
| 176 |
+
],
|
| 177 |
+
"pca_unwrapped_3_neighbor_mean": [
|
| 178 |
+
-0.2933238636363636,
|
| 179 |
+
-0.2173753218217329,
|
| 180 |
+
-0.3480571400035511
|
| 181 |
+
],
|
| 182 |
+
"pca_unwrapped_3_neighbor_std": [
|
| 183 |
+
0.6203425764161097,
|
| 184 |
+
0.5938304683645145,
|
| 185 |
+
0.5600074530240728
|
| 186 |
+
],
|
| 187 |
+
"pca_unwrapped_4": [
|
| 188 |
+
0.67333984375,
|
| 189 |
+
-0.0005474090576171,
|
| 190 |
+
0.81982421875
|
| 191 |
+
],
|
| 192 |
+
"pca_unwrapped_4_neighbor_mean": [
|
| 193 |
+
0.2915762121027166,
|
| 194 |
+
0.3528386896306818,
|
| 195 |
+
0.2782594507390802
|
| 196 |
+
],
|
| 197 |
+
"pca_unwrapped_4_neighbor_std": [
|
| 198 |
+
0.6415192812587974,
|
| 199 |
+
0.6430080201673403,
|
| 200 |
+
0.6308895861182334
|
| 201 |
+
],
|
| 202 |
+
"pca_unwrapped_5": [
|
| 203 |
+
0.73876953125,
|
| 204 |
+
0.50048828125,
|
| 205 |
+
-0.03192138671875
|
| 206 |
+
],
|
| 207 |
+
"pca_unwrapped_5_neighbor_mean": [
|
| 208 |
+
0.2028697620738636,
|
| 209 |
+
0.2245316938920454,
|
| 210 |
+
0.2729325727982954
|
| 211 |
+
],
|
| 212 |
+
"pca_unwrapped_5_neighbor_std": [
|
| 213 |
+
0.265173781606759,
|
| 214 |
+
0.2994363858938455,
|
| 215 |
+
0.2968562365279343
|
| 216 |
+
],
|
| 217 |
+
"pca_unwrapped_6": [
|
| 218 |
+
0.99951171875,
|
| 219 |
+
0.05828857421875,
|
| 220 |
+
-0.77880859375
|
| 221 |
+
],
|
| 222 |
+
"pca_unwrapped_6_neighbor_mean": [
|
| 223 |
+
-0.2386505820534446,
|
| 224 |
+
-0.1530848416415128,
|
| 225 |
+
-0.0769850990988991
|
| 226 |
+
],
|
| 227 |
+
"pca_unwrapped_6_neighbor_std": [
|
| 228 |
+
0.6776577717043619,
|
| 229 |
+
0.7717860533115238,
|
| 230 |
+
0.7447135522384378
|
| 231 |
+
],
|
| 232 |
+
"pca_unwrapped_7": [
|
| 233 |
+
0.023834228515625,
|
| 234 |
+
-0.9931640625,
|
| 235 |
+
0.52978515625
|
| 236 |
+
],
|
| 237 |
+
"pca_unwrapped_7_neighbor_mean": [
|
| 238 |
+
-0.4803272594105113,
|
| 239 |
+
-0.3878728693181818,
|
| 240 |
+
-0.5263227982954546
|
| 241 |
+
],
|
| 242 |
+
"pca_unwrapped_7_neighbor_std": [
|
| 243 |
+
0.4799926318285017,
|
| 244 |
+
0.4691567465869561,
|
| 245 |
+
0.3891669942534205
|
| 246 |
+
],
|
| 247 |
+
"pca_unwrapped_8": [
|
| 248 |
+
0.0192413330078125,
|
| 249 |
+
0.0997314453125,
|
| 250 |
+
-0.3359375
|
| 251 |
+
],
|
| 252 |
+
"pca_unwrapped_8_neighbor_mean": [
|
| 253 |
+
-0.0384375832297585,
|
| 254 |
+
-0.0457548661665482,
|
| 255 |
+
-0.0061485984108664
|
| 256 |
+
],
|
| 257 |
+
"pca_unwrapped_8_neighbor_std": [
|
| 258 |
+
0.3037878488292577,
|
| 259 |
+
0.3010843368506175,
|
| 260 |
+
0.2874409267860334
|
| 261 |
+
],
|
| 262 |
+
"pca_val_unwrapped_0": [
|
| 263 |
+
15657.09765625,
|
| 264 |
+
40668.40625,
|
| 265 |
+
66863.0
|
| 266 |
+
],
|
| 267 |
+
"pca_val_unwrapped_0_neighbor_mean": [
|
| 268 |
+
69378.52059659091,
|
| 269 |
+
67104.76526988637,
|
| 270 |
+
64723.43856534091
|
| 271 |
+
],
|
| 272 |
+
"pca_val_unwrapped_0_neighbor_std": [
|
| 273 |
+
20242.245019019712,
|
| 274 |
+
24702.906417865197,
|
| 275 |
+
25959.16138296664
|
| 276 |
+
],
|
| 277 |
+
"pca_val_unwrapped_1": [
|
| 278 |
+
11305.3017578125,
|
| 279 |
+
34416.42578125,
|
| 280 |
+
59273.25
|
| 281 |
+
],
|
| 282 |
+
"pca_val_unwrapped_1_neighbor_mean": [
|
| 283 |
+
41190.40261008523,
|
| 284 |
+
39089.39133522727,
|
| 285 |
+
36829.68004261364
|
| 286 |
+
],
|
| 287 |
+
"pca_val_unwrapped_1_neighbor_std": [
|
| 288 |
+
16625.870141811894,
|
| 289 |
+
18875.56976212627,
|
| 290 |
+
17666.778281657556
|
| 291 |
+
],
|
| 292 |
+
"pca_val_unwrapped_2": [
|
| 293 |
+
1270.4095458984375,
|
| 294 |
+
13551.6748046875,
|
| 295 |
+
47764.625
|
| 296 |
+
],
|
| 297 |
+
"pca_val_unwrapped_2_neighbor_mean": [
|
| 298 |
+
28717.50048828125,
|
| 299 |
+
27601.021828391335,
|
| 300 |
+
24490.75362881747
|
| 301 |
+
],
|
| 302 |
+
"pca_val_unwrapped_2_neighbor_std": [
|
| 303 |
+
14988.204981576571,
|
| 304 |
+
16601.48080038032,
|
| 305 |
+
15622.078784778376
|
| 306 |
+
],
|
| 307 |
+
"post_synapse_count": [
|
| 308 |
+
0.0,
|
| 309 |
+
0.0,
|
| 310 |
+
0.0
|
| 311 |
+
],
|
| 312 |
+
"post_synapse_count_neighbor_mean": [
|
| 313 |
+
0.0,
|
| 314 |
+
0.0,
|
| 315 |
+
0.0
|
| 316 |
+
],
|
| 317 |
+
"post_synapse_count_neighbor_std": [
|
| 318 |
+
0.0,
|
| 319 |
+
0.0,
|
| 320 |
+
0.0
|
| 321 |
+
],
|
| 322 |
+
"pre_synapse_count": [
|
| 323 |
+
0.0,
|
| 324 |
+
0.0,
|
| 325 |
+
0.0
|
| 326 |
+
],
|
| 327 |
+
"pre_synapse_count_neighbor_mean": [
|
| 328 |
+
0.0,
|
| 329 |
+
0.0,
|
| 330 |
+
0.0
|
| 331 |
+
],
|
| 332 |
+
"pre_synapse_count_neighbor_std": [
|
| 333 |
+
0.0,
|
| 334 |
+
0.0,
|
| 335 |
+
0.0
|
| 336 |
+
],
|
| 337 |
+
"size_nm3": [
|
| 338 |
+
12771840.0,
|
| 339 |
+
697943040.0,
|
| 340 |
+
7550330880.0
|
| 341 |
+
],
|
| 342 |
+
"size_nm3_neighbor_mean": [
|
| 343 |
+
3233702034.285714,
|
| 344 |
+
3184761234.285714,
|
| 345 |
+
2695304960.0
|
| 346 |
+
],
|
| 347 |
+
"size_nm3_neighbor_std": [
|
| 348 |
+
3650678969.7909584,
|
| 349 |
+
3691650923.5639486,
|
| 350 |
+
3518520747.0511127
|
| 351 |
+
]
|
| 352 |
+
},
|
| 353 |
+
"model": {
|
| 354 |
+
"file": "local_compartment_classifier_bd_boxes.skops"
|
| 355 |
+
},
|
| 356 |
+
"model_format": "skops",
|
| 357 |
+
"task": "tabular-classification",
|
| 358 |
+
"use_intelex": false
|
| 359 |
+
}
|
| 360 |
+
}
|
local_compartment_classifier_bd_boxes.skops
ADDED
|
Binary file (515 kB). View file
|
|
|
train.py
ADDED
|
@@ -0,0 +1,359 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# %%
|
| 2 |
+
|
| 3 |
+
|
| 4 |
+
from pathlib import Path
|
| 5 |
+
|
| 6 |
+
import caveclient as cc
|
| 7 |
+
import matplotlib.pyplot as plt
|
| 8 |
+
import numpy as np
|
| 9 |
+
import pandas as pd
|
| 10 |
+
import seaborn as sns
|
| 11 |
+
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
|
| 12 |
+
from sklearn.ensemble import RandomForestClassifier
|
| 13 |
+
from sklearn.metrics import classification_report
|
| 14 |
+
from sklearn.model_selection import KFold
|
| 15 |
+
from sklearn.pipeline import Pipeline
|
| 16 |
+
from sklearn.preprocessing import QuantileTransformer
|
| 17 |
+
from skops.io import dump
|
| 18 |
+
|
| 19 |
+
client = cc.CAVEclient("minnie65_phase3_v1")
|
| 20 |
+
|
| 21 |
+
out_path = Path("./troglobyte-sandbox/models/")
|
| 22 |
+
|
| 23 |
+
model_name = "local_compartment_classifier_bd_boxes"
|
| 24 |
+
|
| 25 |
+
data_path = Path("./troglobyte-sandbox/data/bounding_box_labels")
|
| 26 |
+
|
| 27 |
+
files = list(data_path.glob("*.csv"))
|
| 28 |
+
|
| 29 |
+
# %%
|
| 30 |
+
label_df = pd.read_csv(out_path / model_name / "labels.csv", index_col=[0, 1])
|
| 31 |
+
label_df = label_df.rename(columns=lambda x: x.replace(".1", ""))
|
| 32 |
+
|
| 33 |
+
# # %%
|
| 34 |
+
|
| 35 |
+
# X_df = wrangler.features_.copy()
|
| 36 |
+
# X_df = X_df.drop(columns=[col for col in X_df.columns if "rep_coord" in col])
|
| 37 |
+
|
| 38 |
+
# %%
|
| 39 |
+
|
| 40 |
+
X_df = pd.read_csv(out_path / model_name / "features.csv", index_col=[0, 1])
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
# %%
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
def box_train_test_split(
|
| 47 |
+
train_box_indices, test_box_indices, X_df, label_df, label_column
|
| 48 |
+
):
|
| 49 |
+
train_label_df = label_df.loc[train_box_indices + 1].droplevel("bbox_id")
|
| 50 |
+
test_label_df = label_df.loc[test_box_indices + 1].droplevel("bbox_id")
|
| 51 |
+
|
| 52 |
+
train_X_df = X_df.loc[train_label_df["root_id"]]
|
| 53 |
+
test_X_df = X_df.loc[test_label_df["root_id"]]
|
| 54 |
+
train_X_df = train_X_df.dropna()
|
| 55 |
+
test_X_df = test_X_df.dropna()
|
| 56 |
+
|
| 57 |
+
train_l2_y = train_X_df.index.get_level_values("object_id").map(
|
| 58 |
+
train_label_df[label_column]
|
| 59 |
+
)
|
| 60 |
+
test_l2_y = test_X_df.index.get_level_values("object_id").map(
|
| 61 |
+
test_label_df[label_column]
|
| 62 |
+
)
|
| 63 |
+
|
| 64 |
+
# TODO do something more fair here w/ evaluation on the uncertains
|
| 65 |
+
train_X_df = train_X_df.loc[train_l2_y.notna()]
|
| 66 |
+
train_l2_y = train_l2_y[train_l2_y.notna()].values.astype(str)
|
| 67 |
+
|
| 68 |
+
test_X_df = test_X_df.loc[test_l2_y.notna()]
|
| 69 |
+
test_l2_y = test_l2_y[test_l2_y.notna()].values.astype(str)
|
| 70 |
+
|
| 71 |
+
return train_X_df, test_X_df, train_l2_y, test_l2_y
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
def aggregate_votes_by_object(X_df, l2_node_predictions):
|
| 75 |
+
l2_node_predictions = pd.Series(
|
| 76 |
+
index=X_df.index, data=l2_node_predictions, name="label"
|
| 77 |
+
)
|
| 78 |
+
|
| 79 |
+
object_prediction_counts = (
|
| 80 |
+
l2_node_predictions.groupby(level="object_id").value_counts().to_frame()
|
| 81 |
+
)
|
| 82 |
+
|
| 83 |
+
object_n_predictions = object_prediction_counts.groupby("object_id").sum()
|
| 84 |
+
|
| 85 |
+
sufficient_data_index = object_n_predictions.query("count > 3").index
|
| 86 |
+
|
| 87 |
+
object_prediction_counts = object_prediction_counts.loc[sufficient_data_index]
|
| 88 |
+
|
| 89 |
+
object_prediction_probs = object_prediction_counts.unstack(fill_value=0)
|
| 90 |
+
object_prediction_probs = object_prediction_probs.div(
|
| 91 |
+
object_prediction_probs.sum(axis=1), axis=0
|
| 92 |
+
)
|
| 93 |
+
|
| 94 |
+
object_prediction_counts.reset_index(drop=False, inplace=True)
|
| 95 |
+
|
| 96 |
+
max_locs = object_prediction_counts.groupby("object_id")["count"].idxmax()
|
| 97 |
+
|
| 98 |
+
max_predictions = object_prediction_counts.loc[max_locs]
|
| 99 |
+
max_predictions["proportion"] = (
|
| 100 |
+
max_predictions["count"]
|
| 101 |
+
/ object_n_predictions.loc[max_predictions["object_id"]]["count"].values
|
| 102 |
+
)
|
| 103 |
+
max_predictions = max_predictions.set_index("object_id")
|
| 104 |
+
return max_predictions, object_prediction_probs
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
# models to evaluate
|
| 108 |
+
def get_lda(n_classes):
|
| 109 |
+
lda = Pipeline(
|
| 110 |
+
[
|
| 111 |
+
("transformer", QuantileTransformer(output_distribution="normal")),
|
| 112 |
+
("lda", LinearDiscriminantAnalysis(n_components=n_classes - 1)),
|
| 113 |
+
]
|
| 114 |
+
)
|
| 115 |
+
return lda
|
| 116 |
+
|
| 117 |
+
|
| 118 |
+
rf = RandomForestClassifier(n_estimators=500, max_depth=4)
|
| 119 |
+
|
| 120 |
+
box_indices = np.arange(1, 4)
|
| 121 |
+
|
| 122 |
+
rows = []
|
| 123 |
+
for fold, (train_box_indices, test_box_indices) in enumerate(
|
| 124 |
+
KFold(n_splits=3).split(box_indices.reshape(-1, 1))
|
| 125 |
+
):
|
| 126 |
+
for label_column in ["axon_label", "simple_label"]:
|
| 127 |
+
train_X_df, test_X_df, train_l2_y, test_l2_y = box_train_test_split(
|
| 128 |
+
train_box_indices, test_box_indices, X_df, label_df, label_column
|
| 129 |
+
)
|
| 130 |
+
n_classes = label_df[label_column].nunique()
|
| 131 |
+
models = {"rf": rf, "lda": get_lda(n_classes)}
|
| 132 |
+
for model_type, model in models.items():
|
| 133 |
+
model.fit(train_X_df, train_l2_y)
|
| 134 |
+
train_preds = model.predict(train_X_df)
|
| 135 |
+
test_preds = model.predict(test_X_df)
|
| 136 |
+
|
| 137 |
+
# evaluate at the L2 level
|
| 138 |
+
train_report = classification_report(
|
| 139 |
+
train_l2_y, train_preds, output_dict=True
|
| 140 |
+
)
|
| 141 |
+
rows.append(
|
| 142 |
+
{
|
| 143 |
+
"model": model_type,
|
| 144 |
+
"fold": fold,
|
| 145 |
+
"accuracy": train_report["accuracy"],
|
| 146 |
+
"macro_f1": train_report["macro avg"]["f1-score"],
|
| 147 |
+
"weighted_f1": train_report["weighted avg"]["f1-score"],
|
| 148 |
+
"evaluation": "train",
|
| 149 |
+
"labeling": label_column,
|
| 150 |
+
"level": "level2",
|
| 151 |
+
}
|
| 152 |
+
)
|
| 153 |
+
|
| 154 |
+
test_report = classification_report(test_l2_y, test_preds, output_dict=True)
|
| 155 |
+
rows.append(
|
| 156 |
+
{
|
| 157 |
+
"model": model_type,
|
| 158 |
+
"fold": fold,
|
| 159 |
+
"accuracy": test_report["accuracy"],
|
| 160 |
+
"macro_f1": test_report["macro avg"]["f1-score"],
|
| 161 |
+
"weighted_f1": test_report["weighted avg"]["f1-score"],
|
| 162 |
+
"evaluation": "test",
|
| 163 |
+
"labeling": label_column,
|
| 164 |
+
"level": "level2",
|
| 165 |
+
}
|
| 166 |
+
)
|
| 167 |
+
|
| 168 |
+
# evaluate at the object level
|
| 169 |
+
train_object_predictions, train_object_probs = aggregate_votes_by_object(
|
| 170 |
+
train_X_df, train_preds
|
| 171 |
+
)
|
| 172 |
+
train_object_y = (
|
| 173 |
+
label_df.droplevel(0)
|
| 174 |
+
.loc[train_object_predictions.index, label_column]
|
| 175 |
+
.values.astype(str)
|
| 176 |
+
)
|
| 177 |
+
train_object_report = classification_report(
|
| 178 |
+
train_object_y, train_object_predictions["label"], output_dict=True
|
| 179 |
+
)
|
| 180 |
+
rows.append(
|
| 181 |
+
{
|
| 182 |
+
"model": model_type + "-vote",
|
| 183 |
+
"fold": fold,
|
| 184 |
+
"accuracy": train_object_report["accuracy"],
|
| 185 |
+
"macro_f1": train_object_report["macro avg"]["f1-score"],
|
| 186 |
+
"weighted_f1": train_object_report["weighted avg"]["f1-score"],
|
| 187 |
+
"evaluation": "train",
|
| 188 |
+
"labeling": label_column,
|
| 189 |
+
"level": "root",
|
| 190 |
+
}
|
| 191 |
+
)
|
| 192 |
+
|
| 193 |
+
test_object_predictions, test_object_probs = aggregate_votes_by_object(
|
| 194 |
+
test_X_df, test_preds
|
| 195 |
+
)
|
| 196 |
+
test_object_y = (
|
| 197 |
+
label_df.droplevel(0)
|
| 198 |
+
.loc[test_object_predictions.index, label_column]
|
| 199 |
+
.values.astype(str)
|
| 200 |
+
)
|
| 201 |
+
test_object_report = classification_report(
|
| 202 |
+
test_object_y, test_object_predictions["label"], output_dict=True
|
| 203 |
+
)
|
| 204 |
+
rows.append(
|
| 205 |
+
{
|
| 206 |
+
"model": model_type + "-vote",
|
| 207 |
+
"fold": fold,
|
| 208 |
+
"accuracy": test_object_report["accuracy"],
|
| 209 |
+
"macro_f1": train_object_report["macro avg"]["f1-score"],
|
| 210 |
+
"weighted_f1": train_object_report["weighted avg"]["f1-score"],
|
| 211 |
+
"evaluation": "test",
|
| 212 |
+
"labeling": label_column,
|
| 213 |
+
"level": "root",
|
| 214 |
+
}
|
| 215 |
+
)
|
| 216 |
+
|
| 217 |
+
|
| 218 |
+
# %%
|
| 219 |
+
|
| 220 |
+
evaluation_df = pd.DataFrame(rows)
|
| 221 |
+
|
| 222 |
+
sns.set_context("talk")
|
| 223 |
+
|
| 224 |
+
fig, axs = plt.subplots(2, 3, figsize=(15, 10), constrained_layout=True, sharey="col")
|
| 225 |
+
for i, labeling in enumerate(["simple_label", "axon_label"]):
|
| 226 |
+
for j, metric in enumerate(["accuracy", "weighted_f1", "macro_f1"]):
|
| 227 |
+
ax = axs[i, j]
|
| 228 |
+
show_legend = (i == 0) & (j == 0)
|
| 229 |
+
sns.stripplot(
|
| 230 |
+
data=evaluation_df.query("labeling == @labeling"),
|
| 231 |
+
x="model",
|
| 232 |
+
y=metric,
|
| 233 |
+
hue="evaluation",
|
| 234 |
+
ax=ax,
|
| 235 |
+
legend=show_legend,
|
| 236 |
+
s=10,
|
| 237 |
+
jitter=True,
|
| 238 |
+
)
|
| 239 |
+
ax.spines[["right", "top"]].set_visible(False)
|
| 240 |
+
if j == 1:
|
| 241 |
+
ax.set_title("Labeling: " + labeling)
|
| 242 |
+
|
| 243 |
+
|
| 244 |
+
# %%
|
| 245 |
+
lda = model
|
| 246 |
+
train_X_transformed = lda.transform(train_X_df)
|
| 247 |
+
|
| 248 |
+
# %%
|
| 249 |
+
fig, ax = plt.subplots(1, 1, figsize=(10, 10))
|
| 250 |
+
sns.scatterplot(
|
| 251 |
+
x=train_X_transformed[:, 0],
|
| 252 |
+
y=train_X_transformed[:, 1],
|
| 253 |
+
hue=train_l2_y,
|
| 254 |
+
ax=ax,
|
| 255 |
+
s=10,
|
| 256 |
+
alpha=0.7,
|
| 257 |
+
)
|
| 258 |
+
ax.set(xticks=[], yticks=[], xlabel="LDA1", ylabel="LDA2")
|
| 259 |
+
ax.spines[["right", "top"]].set_visible(False)
|
| 260 |
+
# %%
|
| 261 |
+
fig, ax = plt.subplots(1, 1, figsize=(10, 10))
|
| 262 |
+
sns.scatterplot(
|
| 263 |
+
x=train_X_transformed[:, 0],
|
| 264 |
+
y=train_X_transformed[:, 2],
|
| 265 |
+
hue=train_l2_y,
|
| 266 |
+
ax=ax,
|
| 267 |
+
s=10,
|
| 268 |
+
alpha=0.7,
|
| 269 |
+
)
|
| 270 |
+
ax.set(xticks=[], yticks=[], xlabel="LDA1", ylabel="LDA3")
|
| 271 |
+
ax.spines[["right", "top"]].set_visible(False)
|
| 272 |
+
|
| 273 |
+
|
| 274 |
+
# %%
|
| 275 |
+
final_lda = Pipeline(
|
| 276 |
+
[
|
| 277 |
+
("transformer", QuantileTransformer(output_distribution="normal")),
|
| 278 |
+
("lda", LinearDiscriminantAnalysis(n_components=n_classes - 1)),
|
| 279 |
+
]
|
| 280 |
+
)
|
| 281 |
+
|
| 282 |
+
train_X_df, test_X_df, train_l2_y, test_l2_y = box_train_test_split(
|
| 283 |
+
np.array([0, 1, 2]), np.array([]), X_df, label_df, label_column
|
| 284 |
+
)
|
| 285 |
+
|
| 286 |
+
final_lda.fit(train_X_df, train_l2_y)
|
| 287 |
+
|
| 288 |
+
# %%
|
| 289 |
+
|
| 290 |
+
model_pickle_file = out_path / model_name / f"{model_name}.skops"
|
| 291 |
+
with open(model_pickle_file, mode="bw") as f:
|
| 292 |
+
dump(final_lda, file=f)
|
| 293 |
+
|
| 294 |
+
# %%
|
| 295 |
+
from pathlib import Path
|
| 296 |
+
|
| 297 |
+
from skops import card, hub_utils
|
| 298 |
+
|
| 299 |
+
hub_out_path = Path(
|
| 300 |
+
"troglobyte-sandbox/models/local_compartment_classifier_bd_boxes/hub"
|
| 301 |
+
)
|
| 302 |
+
if not hub_out_path.exists():
|
| 303 |
+
hub_utils.init(
|
| 304 |
+
model=model_pickle_file,
|
| 305 |
+
requirements=["scikit-learn", "caveclient"],
|
| 306 |
+
dst=hub_out_path,
|
| 307 |
+
task="tabular-classification",
|
| 308 |
+
data=train_X_df,
|
| 309 |
+
)
|
| 310 |
+
|
| 311 |
+
hub_utils.add_files(__file__, dst=hub_out_path, exist_ok=True)
|
| 312 |
+
|
| 313 |
+
model_card = card.Card(model, metadata=card.metadata_from_config(hub_out_path))
|
| 314 |
+
|
| 315 |
+
model_card.metadata.license = "mit"
|
| 316 |
+
|
| 317 |
+
model_description = (
|
| 318 |
+
"This is a model trained to classify pieces of neuron as axon, dendrite, soma, or"
|
| 319 |
+
"glia, "
|
| 320 |
+
"based only on their local shape and synapse features."
|
| 321 |
+
"The model is a linear discriminant classifier which was trained on compartment "
|
| 322 |
+
"labels generated by Bethanny Danskin for 3 6x6x6 um boxes in the Minnie65 Phase3 "
|
| 323 |
+
"dataset."
|
| 324 |
+
)
|
| 325 |
+
|
| 326 |
+
model_card_authors = "bdpedigo"
|
| 327 |
+
|
| 328 |
+
model_card.add(
|
| 329 |
+
model_card_authors=model_card_authors,
|
| 330 |
+
model_description=model_description,
|
| 331 |
+
)
|
| 332 |
+
|
| 333 |
+
model_card.save(hub_out_path / "README.md")
|
| 334 |
+
|
| 335 |
+
hub_utils.push(
|
| 336 |
+
repo_id=f"bdpedigo/{model_name}",
|
| 337 |
+
source=hub_out_path,
|
| 338 |
+
create_remote=False,
|
| 339 |
+
private=False,
|
| 340 |
+
)
|
| 341 |
+
|
| 342 |
+
# %%
|
| 343 |
+
|
| 344 |
+
syn_features = [col for col in X_df.columns if "syn" in col]
|
| 345 |
+
train_X_df_no_syn = train_X_df.drop(columns=syn_features)
|
| 346 |
+
|
| 347 |
+
final_lda_no_syn = Pipeline(
|
| 348 |
+
[
|
| 349 |
+
("transformer", QuantileTransformer(output_distribution="normal")),
|
| 350 |
+
("lda", LinearDiscriminantAnalysis(n_components=n_classes - 1)),
|
| 351 |
+
]
|
| 352 |
+
)
|
| 353 |
+
|
| 354 |
+
final_lda_no_syn.fit(train_X_df_no_syn, train_l2_y)
|
| 355 |
+
|
| 356 |
+
print(classification_report(train_l2_y, final_lda_no_syn.predict(train_X_df_no_syn)))
|
| 357 |
+
|
| 358 |
+
with open(out_path / model_name / f"{model_name}_no_syn.skops", mode="bw") as f:
|
| 359 |
+
dump(final_lda_no_syn, file=f)
|