super optimizing MVE, v7, v8 running

deepskies · Aug 4, 2024 · 6e18b24 · 6e18b24
1 parent c2e166a
commit 6e18b24
Show file tree

Hide file tree

Showing 94 changed files with 18,734 additions and 1,194 deletions.
diff --git a/src/training/MVE/MVE_SL_DA_v3_attempt2.ipynb b/src/training/MVE/MVE_SL_DA_v3_attempt2.ipynb
diff --git a/src/training/MVE/MVE_SL_DA_v4.ipynb b/src/training/MVE/MVE_SL_DA_v4.ipynb
diff --git a/src/training/MVE/MVE_SL_DA_v5.ipynb b/src/training/MVE/MVE_SL_DA_v5.ipynb
diff --git a/src/training/MVE/MVE_SL_DA_v6.ipynb b/src/training/MVE/MVE_SL_DA_v6.ipynb
diff --git a/src/training/MVE/MVE_SL_DA_v7.ipynb b/src/training/MVE/MVE_SL_DA_v7.ipynb
diff --git a/src/training/MVE/MVE_SL_DA_v8.ipynb b/src/training/MVE/MVE_SL_DA_v8.ipynb
diff --git a/src/training/MVE/README.md b/src/training/MVE/README.md
@@ -2,22 +2,70 @@ MVE_SL_DA_v1:
     DA_weight: 1.4 -> 1.0
     beta: 1.0 -> 0.4 for 75 epochs, then fixed at 0.4s
     lr: 1e-5
+    batch_size: 32
 
 MVE_SL_DA_v2:
     DA_weight: 1.4 -> 1.0
     beta: 1.0 -> 0.4 
     lr: 1e-5
-
+    batch_size: 32
+
 MVE_SL_DA_v3:
     DA_weight: 1.4 -> 1.0
     beta: 1.0 -> 0.4 
     lr: 3e-5
+    batch_size: 32
     comment: suffered some unexpected problem, model froze
 
 MVE_SL_DA_v3_attempt2:
     DA_weight: 1.4 -> 1.0
     beta: 1.0 -> 0.4 
     lr: 3e-5
-    comment: same as v3, trying different seed.
+    batch_size: 32
+    comment: same as v3, trying different seed. Didn't help.
+
+MVE_SL_DA_v4:
+    DA_weight: 1.4 -> 1.0
+    beta: 1.0 -> 0.4 
+    lr: 3e-5
+    batch_size: 64
+    comment: higher batch size with LR expected to make gradient stabler. Great improvement.
+
+MVE_SL_DA_v5:
+    DA_weight: 1.4 -> 1.0
+    beta: 1.0 -> 0.4 
+    lr: 1e-5
+    batch_size: 32
+    comment: added regression layer, check performance improvement or worsens. Uncertainties are much smaller.
+
+MVE_SL_DA_v6:
+    DA_weight: 1.4 -> 1.0
+    beta: 1.0 -> 0.4 
+    lr: 3e-5
+    batch_size: 256
+    comment: go crazy w/ batch size with LR expected to make gradient stabler. Didn't learn variance so well.
+
+MVE_SL_DA_v7:
+    DA_weight: 1.4 -> 1.0
+    beta: 1.0 -> 0.0
+    lr: 3e-5
+    batch_size: 64
+    epochs: 250
+    comment: go crazy w/ batch size with LR expected to make gradient stabler. 
 
+MVE_SL_DA_v8:
+    DA_weight: 1.4 -> 1.0
+    beta: 1.0 -> 0.0
+    lr: 3e-5
+    batch_size: 64
+    epochs: 250
+    comment: Combine knowledge for v1 NN model.
 
+MVE_SL_DA_v9:
+    DA_weight: 1.4 -> 1.0
+    beta: 1.0 -> 0.0
+    lr: 3e-5
+    batch_size: 64
+    epochs: 250
+    regression layer: true
+    comment: Combine knowledge for v2 NN model.
diff --git a/src/training/MVE/VisualizeModel.ipynb b/src/training/MVE/VisualizeModel.ipynb
diff --git a/src/training/MVE/models/mve_v3_aug4_2020_60_isomap.png b/src/training/MVE/models/mve_v3_aug4_2020_60_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2020_mve b/src/training/MVE/models/mve_v3_aug4_2020_mve
diff --git a/src/training/MVE/models/mve_v3_aug4_2020_mve.json b/src/training/MVE/models/mve_v3_aug4_2020_mve.json
diff --git a/src/training/MVE/models/mve_v3_aug4_2300 b/src/training/MVE/models/mve_v3_aug4_2300
diff --git a/src/training/MVE/models/mve_v3_aug4_2300.json b/src/training/MVE/models/mve_v3_aug4_2300.json
@@ -0,0 +1 @@
+{"train_DA_loss": [0.1626009997899833, 0.15166031603975533, 0.14411562293686203, 0.1399387811502983, 0.13749359494289706, 0.13581846234213282, 0.1347698505892941, 0.13406277535793115, 0.13461837151395992, 0.1373553010660617, 0.13816960068109702, 0.1398111357672014, 0.14278513182233848, 0.1449420646350606, 0.1436696788960118, 0.1421849383977693, 0.14183385409346663, 0.14259768777971293, 0.14253880647462291], "train_regression_loss": [0.5957583684671731, 0.14735354466101588, 0.06862734759761058, 0.0477300791522757, 0.04253292468442859, 0.040194702207858264, 0.03817543031058702, 0.035930195439375354, 0.03408926431603825, 0.032120840613759194, 0.030768011364629543, 0.02906427605305773, 0.027968286259961356, 0.027333974463259036, 0.02694845231764114, 0.026702825632967886, 0.026072511865742082, 0.026133840083837564, 0.025656821625051812], "train_mve_loss": [0.1424759550688795, -0.07866252362523807, -0.08017261059089842, -0.13875767581634754, -0.12299043767603704, -0.12886772559141874, -0.14019067475139768, -0.10439802069468951, -0.07509011049947312, -0.058230585433964976, -0.07520976031396599, -0.052829165719500595, -0.05266132603026586, -0.056413434677411695, -0.06386315370305447, -0.06978175576377857, -0.04419747173745301, -0.10043201894351235, -0.049122531869375506], "train_r2_score": [-0.8608092807808783, 0.5391375138692485, 0.783598988479059, 0.8497978053157722, 0.8650193400343339, 0.8727682012693678, 0.8793465927929115, 0.8866167833192298, 0.8919208966899141, 0.8983251202090382, 0.9025396303097712, 0.9083321204487462, 0.9123258389313775, 0.9138069765575957, 0.9149271902533225, 0.9156685825429769, 0.9177916590140156, 0.9174184128618784, 0.9190741262903626], "val_source_regression_loss": [0.22524695440082793, 0.09795768891170525, 0.054007954302297274, 0.04654612848950419, 0.04302244860038256, 0.04079047194830361, 0.03877241939163891, 0.03645399071655835, 0.03484284307119573, 0.033177374672310744, 0.031885301729865895, 0.030133488653287008, 0.029251516114469547, 0.029474291019141674, 0.0293473322063115, 0.02742535629232598, 0.028495858939495057, 0.027071581957683822, 0.027584498012616376], "val_target_regression_loss": [0.22214338914224296, 0.0968635058754189, 0.05552026749274153, 0.04987751399493141, 0.04633491804265672, 0.04401998220687839, 0.04158489485004335, 0.03944026006112812, 0.03714517149243765, 0.034945124285710845, 0.034072547468837276, 0.03217981542181817, 0.03147841769323987, 0.030884087726378896, 0.030299622737535625, 0.029069607395227928, 0.02851237805121264, 0.028319595114677954, 0.02842194903760579], "val_source_r2_score": [0.29613270622176263, 0.6929579725840093, 0.830700381506598, 0.8543490968457937, 0.8637623415541612, 0.8736498665593038, 0.8784225138710382, 0.8850189099698068, 0.8904806795554363, 0.8971553225439491, 0.8986408478185883, 0.9061291241095354, 0.907721181217362, 0.9079111250182557, 0.90861151591353, 0.9152920140372489, 0.9094464099031503, 0.9158490539617236, 0.9130611540286717], "val_target_r2_score": [0.2986545266665705, 0.6935966711749745, 0.8249297019697565, 0.8411840528221459, 0.850586972661334, 0.858755749021125, 0.8665135594267321, 0.8733010807811187, 0.8822273610590626, 0.8883252467908763, 0.8908322797592645, 0.8971214530833183, 0.899076295401363, 0.9023450945982955, 0.9035319337591419, 0.9057143723869899, 0.9093385429976492, 0.9095334374008474, 0.9095742783526637], "val_source_mve_loss": [-0.057409920065313765, -0.06303236303720505, -0.0884768767816246, -0.14991747364876376, -0.09338143292316206, -0.1476551636958578, -0.15862498911702708, -0.0793555307967268, -0.07013756288274837, -0.04858721372712949, -0.05105791939719088, -0.04316362338793126, -0.03518361352372226, -0.04719353219268808, -0.17587823121790674, -0.043900906873546586, -0.04114077494117865, -0.05354067123239967, -0.04148529823513547], "val_target_mve_loss": [-0.060137245732887534, -0.06339160866655741, -0.08626482130330838, -0.1481815456015289, -0.0921330258582428, -0.1448180188608777, -0.15624774441977216, -0.07641657116781374, -0.06844760800243184, -0.046636515670711066, -0.05027697227041053, -0.04140676049646109, -0.03292418362655837, -0.04471804938356208, -0.17539858780089457, -0.04128114143564443, -0.03966538228426769, -0.0513513717017356, -0.039533601006030275], "da_weight": [1.4, 1.3973333333333333, 1.3946666666666665, 1.392, 1.3893333333333333, 1.3866666666666665, 1.384, 1.3813333333333333, 1.3786666666666665, 1.376, 1.3733333333333333, 1.3706666666666665, 1.3679999999999999, 1.3653333333333333, 1.3626666666666667, 1.3599999999999999, 1.3573333333333333, 1.3546666666666667, 1.3519999999999999], "beta": [1.0, 0.996, 0.992, 0.988, 0.984, 0.98, 0.976, 0.972, 0.968, 0.964, 0.96, 0.956, 0.952, 0.948, 0.944, 0.94, 0.9359999999999999, 0.932, 0.928], "epoch_no": 19}
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_0_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_0_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_100_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_100_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_120_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_120_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_140_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_140_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_20_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_20_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_40_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_40_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_60_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_60_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_80_isomap.png b/src/training/MVE/models/mve_v3_aug4_2300_80_isomap.png
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_final b/src/training/MVE/models/mve_v3_aug4_2300_final
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_final.json b/src/training/MVE/models/mve_v3_aug4_2300_final.json
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_mve b/src/training/MVE/models/mve_v3_aug4_2300_mve
diff --git a/src/training/MVE/models/mve_v3_aug4_2300_mve.json b/src/training/MVE/models/mve_v3_aug4_2300_mve.json
@@ -0,0 +1 @@
+{"train_DA_loss": [0.1626009997899833, 0.15166031603975533, 0.14411562293686203, 0.1399387811502983, 0.13749359494289706, 0.13581846234213282, 0.1347698505892941, 0.13406277535793115, 0.13461837151395992, 0.1373553010660617, 0.13816960068109702, 0.1398111357672014, 0.14278513182233848, 0.1449420646350606, 0.1436696788960118], "train_regression_loss": [0.5957583684671731, 0.14735354466101588, 0.06862734759761058, 0.0477300791522757, 0.04253292468442859, 0.040194702207858264, 0.03817543031058702, 0.035930195439375354, 0.03408926431603825, 0.032120840613759194, 0.030768011364629543, 0.02906427605305773, 0.027968286259961356, 0.027333974463259036, 0.02694845231764114], "train_mve_loss": [0.1424759550688795, -0.07866252362523807, -0.08017261059089842, -0.13875767581634754, -0.12299043767603704, -0.12886772559141874, -0.14019067475139768, -0.10439802069468951, -0.07509011049947312, -0.058230585433964976, -0.07520976031396599, -0.052829165719500595, -0.05266132603026586, -0.056413434677411695, -0.06386315370305447], "train_r2_score": [-0.8608092807808783, 0.5391375138692485, 0.783598988479059, 0.8497978053157722, 0.8650193400343339, 0.8727682012693678, 0.8793465927929115, 0.8866167833192298, 0.8919208966899141, 0.8983251202090382, 0.9025396303097712, 0.9083321204487462, 0.9123258389313775, 0.9138069765575957, 0.9149271902533225], "val_source_regression_loss": [0.22524695440082793, 0.09795768891170525, 0.054007954302297274, 0.04654612848950419, 0.04302244860038256, 0.04079047194830361, 0.03877241939163891, 0.03645399071655835, 0.03484284307119573, 0.033177374672310744, 0.031885301729865895, 0.030133488653287008, 0.029251516114469547, 0.029474291019141674, 0.0293473322063115], "val_target_regression_loss": [0.22214338914224296, 0.0968635058754189, 0.05552026749274153, 0.04987751399493141, 0.04633491804265672, 0.04401998220687839, 0.04158489485004335, 0.03944026006112812, 0.03714517149243765, 0.034945124285710845, 0.034072547468837276, 0.03217981542181817, 0.03147841769323987, 0.030884087726378896, 0.030299622737535625], "val_source_r2_score": [0.29613270622176263, 0.6929579725840093, 0.830700381506598, 0.8543490968457937, 0.8637623415541612, 0.8736498665593038, 0.8784225138710382, 0.8850189099698068, 0.8904806795554363, 0.8971553225439491, 0.8986408478185883, 0.9061291241095354, 0.907721181217362, 0.9079111250182557, 0.90861151591353], "val_target_r2_score": [0.2986545266665705, 0.6935966711749745, 0.8249297019697565, 0.8411840528221459, 0.850586972661334, 0.858755749021125, 0.8665135594267321, 0.8733010807811187, 0.8822273610590626, 0.8883252467908763, 0.8908322797592645, 0.8971214530833183, 0.899076295401363, 0.9023450945982955, 0.9035319337591419], "val_source_mve_loss": [-0.057409920065313765, -0.06303236303720505, -0.0884768767816246, -0.14991747364876376, -0.09338143292316206, -0.1476551636958578, -0.15862498911702708, -0.0793555307967268, -0.07013756288274837, -0.04858721372712949, -0.05105791939719088, -0.04316362338793126, -0.03518361352372226, -0.04719353219268808, -0.17587823121790674], "val_target_mve_loss": [-0.060137245732887534, -0.06339160866655741, -0.08626482130330838, -0.1481815456015289, -0.0921330258582428, -0.1448180188608777, -0.15624774441977216, -0.07641657116781374, -0.06844760800243184, -0.046636515670711066, -0.05027697227041053, -0.04140676049646109, -0.03292418362655837, -0.04471804938356208, -0.17539858780089457], "da_weight": [1.4, 1.3973333333333333, 1.3946666666666665, 1.392, 1.3893333333333333, 1.3866666666666665, 1.384, 1.3813333333333333, 1.3786666666666665, 1.376, 1.3733333333333333, 1.3706666666666665, 1.3679999999999999, 1.3653333333333333, 1.3626666666666667], "beta": [1.0, 0.996, 0.992, 0.988, 0.984, 0.98, 0.976, 0.972, 0.968, 0.964, 0.96, 0.956, 0.952, 0.948, 0.944], "epoch_no": 15}
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330 b/src/training/MVE/models/mve_v4_b64_aug4_2330
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330.json b/src/training/MVE/models/mve_v4_b64_aug4_2330.json
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_0_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_0_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_100_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_100_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_120_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_120_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_140_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_140_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_20_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_20_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_40_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_40_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_60_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_60_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_80_isomap.png b/src/training/MVE/models/mve_v4_b64_aug4_2330_80_isomap.png
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_final b/src/training/MVE/models/mve_v4_b64_aug4_2330_final
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_final.json b/src/training/MVE/models/mve_v4_b64_aug4_2330_final.json
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_mve b/src/training/MVE/models/mve_v4_b64_aug4_2330_mve
diff --git a/src/training/MVE/models/mve_v4_b64_aug4_2330_mve.json b/src/training/MVE/models/mve_v4_b64_aug4_2330_mve.json
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000 b/src/training/MVE/models/mve_v5_NNv2_aug5_0000
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000.json b/src/training/MVE/models/mve_v5_NNv2_aug5_0000.json
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_0_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_0_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_100_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_100_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_120_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_120_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_140_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_140_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_20_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_20_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_40_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_40_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_60_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_60_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_80_isomap.png b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_80_isomap.png
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_final b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_final
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_final.json b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_final.json
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_mve b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_mve
diff --git a/src/training/MVE/models/mve_v5_NNv2_aug5_0000_mve.json b/src/training/MVE/models/mve_v5_NNv2_aug5_0000_mve.json
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005 b/src/training/MVE/models/mve_v6_b256_aug4_0005
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005.json b/src/training/MVE/models/mve_v6_b256_aug4_0005.json
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_0_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_0_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_100_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_100_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_120_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_120_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_140_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_140_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_20_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_20_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_40_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_40_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_60_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_60_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_80_isomap.png b/src/training/MVE/models/mve_v6_b256_aug4_0005_80_isomap.png
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_final b/src/training/MVE/models/mve_v6_b256_aug4_0005_final
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_final.json b/src/training/MVE/models/mve_v6_b256_aug4_0005_final.json
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_mve b/src/training/MVE/models/mve_v6_b256_aug4_0005_mve
diff --git a/src/training/MVE/models/mve_v6_b256_aug4_0005_mve.json b/src/training/MVE/models/mve_v6_b256_aug4_0005_mve.json
diff --git a/src/training/MVE/models/mve_v7_aug5_0045 b/src/training/MVE/models/mve_v7_aug5_0045
diff --git a/src/training/MVE/models/mve_v7_aug5_0045.json b/src/training/MVE/models/mve_v7_aug5_0045.json
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_0_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_0_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_100_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_100_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_120_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_120_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_140_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_140_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_160_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_160_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_20_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_20_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_40_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_40_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_60_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_60_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_80_isomap.png b/src/training/MVE/models/mve_v7_aug5_0045_80_isomap.png
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_mve b/src/training/MVE/models/mve_v7_aug5_0045_mve
diff --git a/src/training/MVE/models/mve_v7_aug5_0045_mve.json b/src/training/MVE/models/mve_v7_aug5_0045_mve.json
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040 b/src/training/MVE/models/mve_v8_NNv2_aug5_0040
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040.json b/src/training/MVE/models/mve_v8_NNv2_aug5_0040.json
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_0_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_0_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_100_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_100_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_120_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_120_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_140_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_140_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_160_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_160_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_180_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_180_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_20_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_20_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_40_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_40_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_60_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_60_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_80_isomap.png b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_80_isomap.png
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_mve b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_mve
diff --git a/src/training/MVE/models/mve_v8_NNv2_aug5_0040_mve.json b/src/training/MVE/models/mve_v8_NNv2_aug5_0040_mve.json
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		{"train_DA_loss": [0.1626009997899833, 0.15166031603975533, 0.14411562293686203, 0.1399387811502983, 0.13749359494289706, 0.13581846234213282, 0.1347698505892941, 0.13406277535793115, 0.13461837151395992, 0.1373553010660617, 0.13816960068109702, 0.1398111357672014, 0.14278513182233848, 0.1449420646350606, 0.1436696788960118, 0.1421849383977693, 0.14183385409346663, 0.14259768777971293, 0.14253880647462291], "train_regression_loss": [0.5957583684671731, 0.14735354466101588, 0.06862734759761058, 0.0477300791522757, 0.04253292468442859, 0.040194702207858264, 0.03817543031058702, 0.035930195439375354, 0.03408926431603825, 0.032120840613759194, 0.030768011364629543, 0.02906427605305773, 0.027968286259961356, 0.027333974463259036, 0.02694845231764114, 0.026702825632967886, 0.026072511865742082, 0.026133840083837564, 0.025656821625051812], "train_mve_loss": [0.1424759550688795, -0.07866252362523807, -0.08017261059089842, -0.13875767581634754, -0.12299043767603704, -0.12886772559141874, -0.14019067475139768, -0.10439802069468951, -0.07509011049947312, -0.058230585433964976, -0.07520976031396599, -0.052829165719500595, -0.05266132603026586, -0.056413434677411695, -0.06386315370305447, -0.06978175576377857, -0.04419747173745301, -0.10043201894351235, -0.049122531869375506], "train_r2_score": [-0.8608092807808783, 0.5391375138692485, 0.783598988479059, 0.8497978053157722, 0.8650193400343339, 0.8727682012693678, 0.8793465927929115, 0.8866167833192298, 0.8919208966899141, 0.8983251202090382, 0.9025396303097712, 0.9083321204487462, 0.9123258389313775, 0.9138069765575957, 0.9149271902533225, 0.9156685825429769, 0.9177916590140156, 0.9174184128618784, 0.9190741262903626], "val_source_regression_loss": [0.22524695440082793, 0.09795768891170525, 0.054007954302297274, 0.04654612848950419, 0.04302244860038256, 0.04079047194830361, 0.03877241939163891, 0.03645399071655835, 0.03484284307119573, 0.033177374672310744, 0.031885301729865895, 0.030133488653287008, 0.029251516114469547, 0.029474291019141674, 0.0293473322063115, 0.02742535629232598, 0.028495858939495057, 0.027071581957683822, 0.027584498012616376], "val_target_regression_loss": [0.22214338914224296, 0.0968635058754189, 0.05552026749274153, 0.04987751399493141, 0.04633491804265672, 0.04401998220687839, 0.04158489485004335, 0.03944026006112812, 0.03714517149243765, 0.034945124285710845, 0.034072547468837276, 0.03217981542181817, 0.03147841769323987, 0.030884087726378896, 0.030299622737535625, 0.029069607395227928, 0.02851237805121264, 0.028319595114677954, 0.02842194903760579], "val_source_r2_score": [0.29613270622176263, 0.6929579725840093, 0.830700381506598, 0.8543490968457937, 0.8637623415541612, 0.8736498665593038, 0.8784225138710382, 0.8850189099698068, 0.8904806795554363, 0.8971553225439491, 0.8986408478185883, 0.9061291241095354, 0.907721181217362, 0.9079111250182557, 0.90861151591353, 0.9152920140372489, 0.9094464099031503, 0.9158490539617236, 0.9130611540286717], "val_target_r2_score": [0.2986545266665705, 0.6935966711749745, 0.8249297019697565, 0.8411840528221459, 0.850586972661334, 0.858755749021125, 0.8665135594267321, 0.8733010807811187, 0.8822273610590626, 0.8883252467908763, 0.8908322797592645, 0.8971214530833183, 0.899076295401363, 0.9023450945982955, 0.9035319337591419, 0.9057143723869899, 0.9093385429976492, 0.9095334374008474, 0.9095742783526637], "val_source_mve_loss": [-0.057409920065313765, -0.06303236303720505, -0.0884768767816246, -0.14991747364876376, -0.09338143292316206, -0.1476551636958578, -0.15862498911702708, -0.0793555307967268, -0.07013756288274837, -0.04858721372712949, -0.05105791939719088, -0.04316362338793126, -0.03518361352372226, -0.04719353219268808, -0.17587823121790674, -0.043900906873546586, -0.04114077494117865, -0.05354067123239967, -0.04148529823513547], "val_target_mve_loss": [-0.060137245732887534, -0.06339160866655741, -0.08626482130330838, -0.1481815456015289, -0.0921330258582428, -0.1448180188608777, -0.15624774441977216, -0.07641657116781374, -0.06844760800243184, -0.046636515670711066, -0.05027697227041053, -0.04140676049646109, -0.03292418362655837, -0.04471804938356208, -0.17539858780089457, -0.04128114143564443, -0.03966538228426769, -0.0513513717017356, -0.039533601006030275], "da_weight": [1.4, 1.3973333333333333, 1.3946666666666665, 1.392, 1.3893333333333333, 1.3866666666666665, 1.384, 1.3813333333333333, 1.3786666666666665, 1.376, 1.3733333333333333, 1.3706666666666665, 1.3679999999999999, 1.3653333333333333, 1.3626666666666667, 1.3599999999999999, 1.3573333333333333, 1.3546666666666667, 1.3519999999999999], "beta": [1.0, 0.996, 0.992, 0.988, 0.984, 0.98, 0.976, 0.972, 0.968, 0.964, 0.96, 0.956, 0.952, 0.948, 0.944, 0.94, 0.9359999999999999, 0.932, 0.928], "epoch_no": 19}