Upload ViT model from experiment a3

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+training_curves/ViT_combined_metrics.png filter=lfs diff=lfs merge=lfs -text
+training_notebook_a3.ipynb filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,161 @@
+---
+license: apache-2.0
+tags:
+- vision-transformer
+- image-classification
+- pytorch
+- timm
+- vit
+- gravitational-lensing
+- strong-lensing
+- astronomy
+- astrophysics
+datasets:
+- C21
+metrics:
+- accuracy
+- auc
+- f1
+model-index:
+- name: ViT-a3
+  results:
+  - task:
+      type: image-classification
+      name: Strong Gravitational Lens Discovery
+    dataset:
+      type: common-test-sample
+      name: Common Test Sample (More et al. 2024)
+    metrics:
+    - type: accuracy
+      value: 0.8340
+      name: Average Accuracy
+    - type: auc
+      value: 0.8374
+      name: Average AUC-ROC
+    - type: f1
+      value: 0.5537
+      name: Average F1-Score
+---
+
+# 🌌 vit-gravit-a3
+
+🔭 This model is part of **GraViT**: Transfer Learning with Vision Transformers and MLP-Mixer for Strong Gravitational Lens Discovery
+
+🔗 **GitHub Repository**: [https://github.com/parlange/gravit](https://github.com/parlange/gravit)
+
+## 🛰️ Model Details
+
+- **🤖 Model Type**: ViT
+- **🧪 Experiment**: A3 - C21-all-blocks-ResNet18
+- **🌌 Dataset**: C21
+- **🪐 Fine-tuning Strategy**: all-blocks
+
+
+
+## 💻 Quick Start
+
+```python
+import torch
+import timm
+
+# Load the model directly from the Hub
+model = timm.create_model(
+    'hf-hub:parlange/vit-gravit-a3',
+    pretrained=True
+)
+model.eval()
+
+# Example inference
+dummy_input = torch.randn(1, 3, 224, 224)
+with torch.no_grad():
+    output = model(dummy_input)
+    predictions = torch.softmax(output, dim=1)
+print(f"Lens probability: {predictions[0][1]:.4f}")
+```
+
+## ⚡️ Training Configuration
+
+**Training Dataset:** C21 (Cañameras et al. 2021)  
+**Fine-tuning Strategy:** all-blocks
+
+
+| 🔧 Parameter | 📝 Value |
+|--------------|----------|
+| Batch Size | 192 |
+| Learning Rate | AdamW with ReduceLROnPlateau |
+| Epochs | 100 |
+| Patience | 10 |
+| Optimizer | AdamW |
+| Scheduler | ReduceLROnPlateau |
+| Image Size | 224x224 |
+| Fine Tune Mode | all_blocks |
+| Stochastic Depth Probability | 0.1 |
+
+
+## 📈 Training Curves
+
+![Combined Training Metrics](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/training_curves/ViT_combined_metrics.png)
+
+
+## 🏁 Final Epoch Training Metrics
+
+| Metric | Training | Validation |
+|:---------:|:-----------:|:-------------:|
+| 📉 Loss | 0.0060 | 0.0320 |
+| 🎯 Accuracy | 0.9978 | 0.9950 |
+| 📊 AUC-ROC | 1.0000 | 0.9997 |
+| ⚖️ F1 Score | 0.9978 | 0.9950 |
+
+
+## ☑️ Evaluation Results
+
+### ROC Curves and Confusion Matrices
+
+Performance across all test datasets (a through l) in the Common Test Sample (More et al. 2024):
+
+![ROC + Confusion Matrix - Dataset A](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_a.png)
+![ROC + Confusion Matrix - Dataset B](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_b.png)
+![ROC + Confusion Matrix - Dataset C](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_c.png)
+![ROC + Confusion Matrix - Dataset D](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_d.png)
+![ROC + Confusion Matrix - Dataset E](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_e.png)
+![ROC + Confusion Matrix - Dataset F](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_f.png)
+![ROC + Confusion Matrix - Dataset G](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_g.png)
+![ROC + Confusion Matrix - Dataset H](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_h.png)
+![ROC + Confusion Matrix - Dataset I](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_i.png)
+![ROC + Confusion Matrix - Dataset J](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_j.png)
+![ROC + Confusion Matrix - Dataset K](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_k.png)
+![ROC + Confusion Matrix - Dataset L](https://huggingface.co/parlange/vit-gravit-a3/resolve/main/roc_confusion_matrix/ViT_roc_confusion_matrix_l.png)
+
+### 📋 Performance Summary
+
+Average performance across 12 test datasets from the Common Test Sample (More et al. 2024):
+
+| Metric | Value |
+|-----------|----------|
+| 🎯 Average Accuracy | 0.8340 |
+| 📈 Average AUC-ROC | 0.8374 |
+| ⚖️ Average F1-Score | 0.5537 |
+
+
+## 📘 Citation
+
+If you use this model in your research, please cite:
+
+```bibtex
+@misc{parlange2025gravit,
+      title={GraViT: Transfer Learning with Vision Transformers and MLP-Mixer for Strong Gravitational Lens Discovery}, 
+      author={René Parlange and Juan C. Cuevas-Tello and Octavio Valenzuela and Omar de J. Cabrera-Rosas and Tomás Verdugo and Anupreeta More and Anton T. Jaelani},
+      year={2025},
+      eprint={2509.00226},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2509.00226}, 
+}
+```
+
+---
+
+
+## Model Card Contact
+
+For questions about this model, please contact the author through: https://github.com/parlange/

config.json

@@ -0,0 +1,76 @@
+{
+  "architecture": "vit_base_patch16_224",
+  "num_classes": 2,
+  "num_features": 768,
+  "global_pool": "token",
+  "crop_pct": 0.875,
+  "interpolation": "bicubic",
+  "mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "first_conv": "patch_embed.proj",
+  "classifier": "head",
+  "input_size": [
+    3,
+    224,
+    224
+  ],
+  "pool_size": [
+    7,
+    7
+  ],
+  "pretrained_cfg": {
+    "tag": "gravit_a3",
+    "custom_load": false,
+    "input_size": [
+      3,
+      224,
+      224
+    ],
+    "fixed_input_size": true,
+    "interpolation": "bicubic",
+    "crop_pct": 0.875,
+    "crop_mode": "center",
+    "mean": [
+      0.485,
+      0.456,
+      0.406
+    ],
+    "std": [
+      0.229,
+      0.224,
+      0.225
+    ],
+    "num_classes": 2,
+    "pool_size": [
+      7,
+      7
+    ],
+    "first_conv": "patch_embed.proj",
+    "classifier": "head"
+  },
+  "model_name": "vit_gravit_a3",
+  "experiment": "a3",
+  "training_strategy": "all-blocks",
+  "dataset": "C21",
+  "hyperparameters": {
+    "batch_size": "192",
+    "learning_rate": "AdamW with ReduceLROnPlateau",
+    "epochs": "100",
+    "patience": "10",
+    "optimizer": "AdamW",
+    "scheduler": "ReduceLROnPlateau",
+    "image_size": "224x224",
+    "fine_tune_mode": "all_blocks",
+    "stochastic_depth_probability": "0.1"
+  },
+  "hf_hub_id": "parlange/vit-gravit-a3",
+  "license": "apache-2.0"
+}

evaluation_results.csv

@@ -0,0 +1,145 @@
+Model,Dataset,Loss,Accuracy,AUCROC,F1
+ViT,a,0.3652414279945689,0.8912291732159698,0.8917799263351749,0.4383116883116883
+ViT,b,0.21351260967325467,0.9459289531593839,0.9444456721915285,0.6108597285067874
+ViT,c,0.850089883324636,0.758252121974222,0.8170580110497236,0.2598652550529355
+ViT,d,0.125797027155771,0.9723357434768941,0.9632872928176796,0.7541899441340782
+ViT,e,0.5662868288882347,0.8803512623490669,0.9096268826156058,0.712401055408971
+ViT,f,0.3309973750718474,0.8993106653241422,0.9044572137856804,0.17197452229299362
+ViT,g,0.06683265567384661,0.9775,0.9995110555555556,0.9779303580186366
+ViT,h,0.4043246931489557,0.878,0.9965371111111111,0.8909740840035746
+ViT,i,0.020328776298090816,0.9915,0.999877,0.991546494281452
+ViT,j,7.168746640741825,0.509,0.4538551111111111,0.1088929219600726
+ViT,k,7.1222426953688265,0.523,0.48649888888888887,0.11173184357541899
+ViT,l,2.481225689433318,0.7816614668711332,0.6814674636422873,0.6155851410483195
+MLP-Mixer,a,0.16029433381890995,0.9534737503929582,0.8991058931860036,0.6084656084656085
+MLP-Mixer,b,0.1373700322756056,0.9629047469349261,0.9401123388581952,0.6609195402298851
+MLP-Mixer,c,0.27720538959380253,0.9141779314680918,0.8577974217311233,0.4572564612326044
+MLP-Mixer,d,0.11598697743656425,0.9682489783087079,0.955084714548803,0.6948640483383686
+MLP-Mixer,e,0.40732662754315313,0.8957189901207464,0.9144403239234089,0.7076923076923077
+MLP-Mixer,f,0.10986769875007599,0.9635194795135931,0.9131062483453624,0.3281027104136947
+MLP-Mixer,g,0.028587095644325017,0.9901666666666666,0.9997029999999999,0.9902398676592225
+MLP-Mixer,h,0.10272313961014151,0.9643333333333334,0.998619888888889,0.9654838709677419
+MLP-Mixer,i,0.01725051350519061,0.993,0.9998728888888889,0.9930325149303252
+MLP-Mixer,j,3.9579579369425772,0.5163333333333333,0.6010726111111112,0.09369144284821987
+MLP-Mixer,k,3.9466213275045154,0.5191666666666667,0.6683313333333333,0.09419152276295134
+MLP-Mixer,l,1.3240398510736613,0.8240177674369414,0.744620509879184,0.6619260463226331
+CvT,a,0.2081917969545678,0.9352404904118202,0.9240874769797423,0.5672268907563025
+CvT,b,0.15224700975965233,0.9575605155611443,0.9504567219152855,0.6666666666666666
+CvT,c,0.49155200616401235,0.8597925180760767,0.8735598526703499,0.3770949720670391
+CvT,d,0.09181369168063785,0.9739075762338887,0.9780386740331491,0.7648725212464589
+CvT,e,0.32513720903680565,0.9143798024149287,0.9261106486036479,0.7758620689655172
+CvT,f,0.18988747454602042,0.940825652544342,0.9312245836823443,0.2611218568665377
+CvT,g,0.04789180162362754,0.9845,0.9996506666666666,0.9847165160230074
+CvT,h,0.22777999278716743,0.9326666666666666,0.9980902777777778,0.9368355222013759
+CvT,i,0.01585208964161575,0.9931666666666666,0.9999184444444444,0.9932040444223438
+CvT,j,4.557504334926605,0.5136666666666667,0.4219552222222222,0.1049079754601227
+CvT,k,4.525464609175921,0.5223333333333333,0.7008621111111111,0.10660847880299251
+CvT,l,1.5619914421430425,0.809793242028449,0.7106891758223672,0.6473875110283306
+Swin,a,0.13412107322078123,0.9607041810751336,0.921658379373849,0.6753246753246753
+Swin,b,0.12724596003046465,0.9556743162527507,0.9398747697974218,0.6483790523690773
+Swin,c,0.17965890587556665,0.9443571204023892,0.9029281767955801,0.5949656750572082
+Swin,d,0.06030154695547883,0.9833385727758567,0.9931565377532229,0.8306709265175719
+Swin,e,0.4499740816497384,0.8562019758507134,0.8587376069022933,0.6649616368286445
+Swin,f,0.09911939381673643,0.9672372395631632,0.9347786366220656,0.3806734992679356
+Swin,g,0.038731103701516986,0.985,0.9998738888888888,0.9852216748768473
+Swin,h,0.06651869903318584,0.979,0.999785,0.9794319294809011
+Swin,i,0.0032394034853205087,0.9996666666666667,0.9999967777777777,0.9996667777407531
+Swin,j,5.625512751281262,0.496,0.10115866666666667,0.04182509505703422
+Swin,k,5.590021039650775,0.5106666666666667,0.37088833333333326,0.04302477183833116
+Swin,l,1.8411567582560506,0.822484268415208,0.5854755526940438,0.65565699046056
+CaiT,a,0.20098744187104706,0.9478151524677775,0.9047292817679559,0.6047619047619047
+CaiT,b,0.12978406120776775,0.9685633448601069,0.9467771639042358,0.7175141242937854
+CaiT,c,0.2854890818960897,0.9248663942156554,0.8821031307550645,0.5152129817444219
+CaiT,d,0.10324450797222483,0.9789374410562716,0.9716519337016575,0.7912772585669782
+CaiT,e,0.4533339374577044,0.8990120746432492,0.9009763112086581,0.7341040462427746
+CaiT,f,0.11921021888981406,0.9653009062040121,0.9248623124563286,0.36182336182336183
+CaiT,g,0.02093089486740064,0.9923333333333333,0.9999474444444445,0.9923916639100232
+CaiT,h,0.10348050882376265,0.9691666666666666,0.9995327777777778,0.9700889248181084
+CaiT,i,0.006860514354542829,0.9978333333333333,0.9999788888888889,0.9978380176284717
+CaiT,j,4.86097429022938,0.5145,0.5177155,0.08367411135577225
+CaiT,k,4.8469039183312566,0.52,0.7220807222222222,0.0845518118245391
+CaiT,l,1.617374696626156,0.8247051980328909,0.7258400455085416,0.6629384850025419
+DeiT,a,0.27381800207860135,0.8953159383841559,0.8986252302025783,0.43463497453310695
+DeiT,b,0.1319006348473214,0.9534737503929582,0.9522191528545119,0.6336633663366337
+DeiT,c,0.5039299008644665,0.7972335743476894,0.8374493554327808,0.28412874583795783
+DeiT,d,0.07908134242881022,0.9742219427852876,0.9686187845303867,0.757396449704142
+DeiT,e,0.3087258418422892,0.8869374313940724,0.9196775902520246,0.713091922005571
+DeiT,f,0.2156604077872147,0.9148787855317171,0.9145406292179694,0.18892988929889298
+DeiT,g,0.04722818533703685,0.9821666666666666,0.9995671111111111,0.9824099950682229
+DeiT,h,0.24446571580693124,0.8993333333333333,0.9973094444444446,0.9082066869300912
+DeiT,i,0.01922516017779708,0.9931666666666666,0.9998760000000001,0.9931859730762839
+DeiT,j,3.201324864923954,0.5038333333333334,0.5165665555555555,0.07345160286336756
+DeiT,k,3.173321856930852,0.5148333333333334,0.5144209444444445,0.074992055926279
+DeiT,l,1.1505240741554759,0.7888530484902967,0.6992895835538881,0.6182965299684543
+DeiT3,a,0.14936432559529178,0.9519019176359635,0.9349825046040517,0.6240786240786241
+DeiT3,b,0.14933169767358925,0.9515875510845646,0.9439152854511971,0.6225490196078431
+DeiT3,c,0.22061521056612046,0.9368123231688148,0.9148646408839779,0.5582417582417583
+DeiT3,d,0.10169743415564346,0.9682489783087079,0.9643388581952119,0.7154929577464789
+DeiT3,e,0.49460477683729975,0.8518111964873765,0.8834708241882994,0.6529562982005142
+DeiT3,f,0.13154652377257423,0.9591046394547286,0.9363108852365101,0.3248081841432225
+DeiT3,g,0.05130978459212929,0.9823333333333333,0.9997385555555556,0.982605841811618
+DeiT3,h,0.08910193234775216,0.9745,0.9995217222222222,0.975085490962384
+DeiT3,i,0.02605568784568459,0.9911666666666666,0.9998692222222222,0.9912266181095845
+DeiT3,j,3.271038075208664,0.5056666666666667,0.4370591666666666,0.08286951144094001
+DeiT3,k,3.245784008204937,0.5145,0.5090863888888889,0.08425023577491354
+DeiT3,l,1.112910136327876,0.8202104595209138,0.7017805623573649,0.656912209889001
+Twins_SVT,a,0.18666393640786363,0.9355548569632192,0.9178057090239411,0.5665961945031712
+Twins_SVT,b,0.12907026239917624,0.9559886828041496,0.958646408839779,0.6568627450980392
+Twins_SVT,c,0.37308393637551635,0.8491040553285131,0.8678618784530387,0.3582887700534759
+Twins_SVT,d,0.07745791368212238,0.9789374410562716,0.982377532228361,0.8
+Twins_SVT,e,0.46567638177510545,0.8572996706915478,0.8989404374479679,0.6733668341708543
+Twins_SVT,f,0.1620182347072841,0.9354039191387189,0.9297958448525039,0.24319419237749546
+Twins_SVT,g,0.03985891605913639,0.9833333333333333,0.9995775555555556,0.9835688465330266
+Twins_SVT,h,0.16922680978477,0.9266666666666666,0.9980967777777777,0.9315281668222845
+Twins_SVT,i,0.012495769090950489,0.9955,0.9999184444444444,0.9955097289206719
+Twins_SVT,j,5.5489834444224835,0.49816666666666665,0.40547916666666667,0.051653543307086616
+Twins_SVT,k,5.521620307348669,0.5103333333333333,0.5116922222222222,0.052869116698903935
+Twins_SVT,l,1.8607527904559573,0.8012267992173867,0.6587701738584388,0.6306377124889456
+Twins_PCPVT,a,0.5452342244934938,0.7969192077962904,0.895244014732965,0.3228511530398323
+Twins_PCPVT,b,0.33666214395135274,0.8836843759823955,0.926718232044199,0.45427728613569324
+Twins_PCPVT,c,0.8452249476460681,0.680289217227287,0.8652670349907919,0.23245283018867924
+Twins_PCPVT,d,0.14044462036633035,0.9487582521219742,0.9665690607734807,0.6539278131634819
+Twins_PCPVT,e,0.9610598787387299,0.6893523600439078,0.849345341708923,0.5211505922165821
+Twins_PCPVT,f,0.4996267252911643,0.8163581442181086,0.909773537083411,0.11496827174318776
+Twins_PCPVT,g,0.16858429829776286,0.9403333333333334,0.9974785,0.9434260429835651
+Twins_PCPVT,h,0.4382073585242033,0.8325,0.9937448888888889,0.8559139784946237
+Twins_PCPVT,i,0.06455629007518292,0.9748333333333333,0.999179,0.97533082829603
+Twins_PCPVT,j,2.452636483669281,0.5051666666666667,0.3821618888888889,0.20123755716976055
+Twins_PCPVT,k,2.348608487725258,0.5396666666666666,0.6640067222222222,0.21310541310541312
+Twins_PCPVT,l,1.069163286557427,0.7349690656231823,0.6925868024466941,0.5836517693969098
+PiT,a,1.4542810870805876,0.7161270040867652,0.9052854511970534,0.26645004061738425
+PiT,b,0.48083062532262133,0.8921722728701666,0.9610782688766114,0.488822652757079
+PiT,c,3.0596617273248543,0.5278214397988054,0.823461325966851,0.17923497267759564
+PiT,d,0.0542862427054065,0.9864822382898459,0.9955837937384899,0.8840970350404312
+PiT,e,1.2118028251844757,0.7727771679473107,0.9089305986528419,0.6130841121495327
+PiT,f,1.3020594693266185,0.7730617303074897,0.9206398944503998,0.10067526089625538
+PiT,g,0.2426671743527986,0.9443333333333334,0.9978501666666667,0.9471351693573916
+PiT,h,1.609877428545151,0.7511666666666666,0.97719,0.8003209843520128
+PiT,i,0.01652756105083972,0.9943333333333333,0.999909,0.9943502824858758
+PiT,j,4.793344738483429,0.4985,0.4088527777777778,0.1740323908866319
+PiT,k,4.567205076335464,0.5485,0.8206032222222223,0.18965001495662578
+PiT,l,2.3343544872733246,0.7027655861667812,0.7048351046138528,0.5527174345508077
+ResNet-18,a,1.1835124935604648,0.6365922665828356,0.9285580110497238,0.22830440587449932
+ResNet-18,b,1.0141342383478842,0.7117258723671801,0.936340699815838,0.2716441620333598
+ResNet-18,c,1.8204121201415364,0.5328513046211883,0.8883425414364641,0.18708971553610504
+ResNet-18,d,0.011235734144227473,0.9949701351776171,0.9996408839779005,0.9553072625698324
+ResNet-18,e,1.056551858831839,0.6125137211855104,0.9263868916975707,0.4920863309352518
+ResNet-18,f,1.0591482175922742,0.6988614359848192,0.9375419355678716,0.08085106382978724
+ResNet-18,g,0.5340945276358107,0.8486666666666667,0.9978233333333334,0.8685201274254272
+ResNet-18,h,0.9615578431227186,0.7538333333333334,0.9955927777777778,0.8024080267558529
+ResNet-18,i,0.002392592921940377,0.9988333333333334,0.9999976666666667,0.998834304746045
+ResNet-18,j,7.01291295003891,0.35383333333333333,0.07525633333333334,0.01524003048006096
+ResNet-18,k,6.481211027059704,0.504,0.7462510555555556,0.019762845849802372
+ResNet-18,l,2.7791932088225773,0.6373010417217493,0.6165565470444929,0.4826910023380345
+Ensemble,a,,0.9487582521219742,0.9267523020257826,0.6320541760722348
+Ensemble,b,,0.9698208110657026,0.9539668508287292,0.7446808510638298
+Ensemble,c,,0.8924866394215656,0.8899631675874772,0.45016077170418006
+Ensemble,d,,0.9823954731216599,0.9802780847145488,0.8333333333333334
+Ensemble,e,,0.9023051591657519,0.9130326193899946,0.7588075880758808
+Ensemble,f,,0.9549221593989621,0.9363232543302679,0.3248259860788863
+Ensemble,g,,0.9905,0.999917,0.9905831818932761
+Ensemble,h,,0.9495,0.9994414444444444,0.9518971265280203
+Ensemble,i,,0.9971666666666666,0.9999844444444443,0.997172792283386
+Ensemble,j,,0.5115,0.3640104444444444,0.07801195344447939
+Ensemble,k,,0.5181666666666667,0.6087856666666667,0.07900605288308378
+Ensemble,l,,0.817037702924224,0.6790663441280981,0.6534455128205128

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23ca3e493402ce2d2572f804e125aa63cba69ddd2a333021a823d8821ba61a47
+size 343214864

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61688f57296e5e0fde9bffa9e785857982cebffd7ceae916a938bd946ac0d1d8
+size 343259038

training_curves/ViT_metrics.csv

@@ -0,0 +1,41 @@
+epoch,train_loss,val_loss,train_accuracy,val_accuracy,train_auc,val_auc,train_f1,val_f1
+1,0.1594870525151491,0.05837649315595627,0.94505,0.974,0.985395874375,0.99834,0.9451637851458224,0.9736842105263158
+2,0.04891255444511771,0.04843974649906158,0.9819125,0.984,0.99838926375,0.998526,0.9819009618631878,0.9840637450199203
+3,0.04098877249099314,0.06516808295249939,0.985425,0.975,0.998859596875,0.997988,0.9854198969639374,0.9749247743229689
+4,0.03653456708919257,0.04256650495901704,0.9867375,0.986,0.9990655153125,0.9992679999999999,0.9867379973251003,0.9858585858585859
+5,0.03568214138075709,0.05940364444255829,0.9868,0.979,0.9991360790624999,0.9983299999999999,0.9868,0.9787234042553191
+6,0.029999172747135164,0.03990015828609467,0.9892375,0.988,0.999345638125,0.998996,0.9892325200405188,0.9879759519038076
+7,0.03169462961144745,0.04534034144878388,0.988275,0.988,0.9993018990624999,0.9993799999999999,0.9882685477012357,0.9879275653923542
+8,0.029803055073274298,0.04331541895493865,0.9892625,0.983,0.999391361875,0.99799,0.989260486341189,0.9829145728643216
+9,0.02843129145205021,0.04371415561437607,0.9896125,0.985,0.999460985,0.9982759999999999,0.9896118507406713,0.9850746268656716
+10,0.027733767334651202,0.03946993482299149,0.9898625,0.988,0.9994940365625,0.998422,0.989861612891128,0.9879759519038076
+11,0.026831346929166465,0.054907928824424745,0.9900875,0.982,0.99951548375,0.998138,0.9900891105195406,0.9820717131474104
+12,0.02800533079933375,0.050178519189357756,0.99005,0.982,0.9994389078124999,0.9985700000000001,0.990046018407363,0.9820717131474104
+13,0.025427712446730585,0.0769548636674881,0.9906875,0.977,0.9995551875000002,0.9972219999999999,0.9906887802927098,0.977205153617443
+14,0.025302810053154826,0.052582308113574984,0.9909125,0.981,0.9995339706249999,0.998996,0.9909114775412234,0.9809428284854563
+15,0.026751336156297476,0.043837911561131475,0.9904125,0.987,0.999521585625,0.998282,0.990409502969678,0.986986986986987
+16,0.024374210640136153,0.05087221178412438,0.9911125,0.983,0.999586088125,0.9994480000000001,0.9911117222756991,0.9827411167512691
+17,0.014263770198030398,0.03483956679701805,0.9949125,0.988,0.9998447559375,0.999428,0.9949105278295339,0.9880239520958084
+18,0.01196307003966067,0.032166809320508034,0.995475,0.987,0.999904258125,0.999582,0.9954736420926278,0.9870388833499502
+19,0.011337511599739083,0.027693722724914552,0.9956875,0.991,0.9999116637500001,0.999648,0.9956874460930761,0.9909729187562688
+20,0.010317483875318431,0.026324373602867126,0.9964125,0.992,0.9999230290625,0.999668,0.9964117374942175,0.992
+21,0.009391166536684613,0.031228957712650298,0.99665,0.988,0.999930400625,0.999572,0.9966499162479062,0.9879759519038076
+22,0.00938724503798876,0.03080810856842436,0.9965875,0.989,0.9999409593750002,0.9996319999999999,0.9965871160505557,0.988988988988989
+23,0.00925770991302561,0.036078432086855176,0.99675,0.989,0.9999399396875,0.9995940000000001,0.9967496749674968,0.9890329012961117
+24,0.008796912119051557,0.030019028067586988,0.9971125,0.988,0.9999192143750002,0.9997039999999999,0.9971122473216406,0.9880478087649402
+25,0.009039205305592622,0.03161667108512484,0.99685,0.992,0.9999435053125,0.999608,0.9968496062007751,0.9920159680638723
+26,0.00832465733307763,0.03130640342843253,0.9968125,0.989,0.9999427559375,0.999682,0.9968128585534127,0.988988988988989
+27,0.007553178189964092,0.02917952951043844,0.9972,0.99,0.9999617928124999,0.99969,0.9972,0.9899799599198397
+28,0.007608072868280578,0.028727686047554015,0.99705,0.991,0.9999638615625001,0.999696,0.9970501474926253,0.990990990990991
+29,0.007120153656997718,0.029447013556957245,0.9973375,0.991,0.9999664453124999,0.999688,0.9973374001525057,0.990990990990991
+30,0.007157817871007137,0.02913126605004072,0.9975625,0.991,0.9999617618750001,0.999682,0.9975616160858312,0.990990990990991
+31,0.007336907451570732,0.028489415229298175,0.9971875,0.991,0.9999640784374999,0.999708,0.9971867615249003,0.991008991008991
+32,0.006501691047073109,0.03007084161043167,0.9977375,0.992,0.999968305,0.999672,0.9977373020139262,0.992
+33,0.00693205452451366,0.02848208716508816,0.9975,0.993,0.9999662856249999,0.999718,0.9974997499749975,0.993006993006993
+34,0.007461001089593629,0.028252465724945067,0.997175,0.992,0.9999625840624999,0.99971,0.997174576186428,0.992
+35,0.00710381545725686,0.029008929789066315,0.9974125,0.993,0.9999660171875,0.9997199999999999,0.9974120147527662,0.992992992992993
+36,0.006280751934862929,0.030127739772200585,0.9976625,0.992,0.999974466875,0.9997020000000001,0.9976623538971185,0.9919839679358717
+37,0.006127115182683338,0.03197662199282786,0.9976625,0.991,0.9999751071874999,0.99969,0.9976624123404627,0.990990990990991
+38,0.00666614195129805,0.030885459803044794,0.9975,0.992,0.9999712131249999,0.999694,0.9974996249437416,0.992
+39,0.007003229365494917,0.030323315382003783,0.9974375,0.992,0.9999688965624999,0.999702,0.9974374679683496,0.992
+40,0.006024414020474069,0.03197560523450375,0.9977875,0.995,0.9999746728125001,0.999688,0.997786586967124,0.995004995004995

training_metrics.csv

@@ -0,0 +1,41 @@
+epoch,train_loss,val_loss,train_accuracy,val_accuracy,train_auc,val_auc,train_f1,val_f1
+1,0.1594870525151491,0.05837649315595627,0.94505,0.974,0.985395874375,0.99834,0.9451637851458224,0.9736842105263158
+2,0.04891255444511771,0.04843974649906158,0.9819125,0.984,0.99838926375,0.998526,0.9819009618631878,0.9840637450199203
+3,0.04098877249099314,0.06516808295249939,0.985425,0.975,0.998859596875,0.997988,0.9854198969639374,0.9749247743229689
+4,0.03653456708919257,0.04256650495901704,0.9867375,0.986,0.9990655153125,0.9992679999999999,0.9867379973251003,0.9858585858585859
+5,0.03568214138075709,0.05940364444255829,0.9868,0.979,0.9991360790624999,0.9983299999999999,0.9868,0.9787234042553191
+6,0.029999172747135164,0.03990015828609467,0.9892375,0.988,0.999345638125,0.998996,0.9892325200405188,0.9879759519038076
+7,0.03169462961144745,0.04534034144878388,0.988275,0.988,0.9993018990624999,0.9993799999999999,0.9882685477012357,0.9879275653923542
+8,0.029803055073274298,0.04331541895493865,0.9892625,0.983,0.999391361875,0.99799,0.989260486341189,0.9829145728643216
+9,0.02843129145205021,0.04371415561437607,0.9896125,0.985,0.999460985,0.9982759999999999,0.9896118507406713,0.9850746268656716
+10,0.027733767334651202,0.03946993482299149,0.9898625,0.988,0.9994940365625,0.998422,0.989861612891128,0.9879759519038076
+11,0.026831346929166465,0.054907928824424745,0.9900875,0.982,0.99951548375,0.998138,0.9900891105195406,0.9820717131474104
+12,0.02800533079933375,0.050178519189357756,0.99005,0.982,0.9994389078124999,0.9985700000000001,0.990046018407363,0.9820717131474104
+13,0.025427712446730585,0.0769548636674881,0.9906875,0.977,0.9995551875000002,0.9972219999999999,0.9906887802927098,0.977205153617443
+14,0.025302810053154826,0.052582308113574984,0.9909125,0.981,0.9995339706249999,0.998996,0.9909114775412234,0.9809428284854563
+15,0.026751336156297476,0.043837911561131475,0.9904125,0.987,0.999521585625,0.998282,0.990409502969678,0.986986986986987
+16,0.024374210640136153,0.05087221178412438,0.9911125,0.983,0.999586088125,0.9994480000000001,0.9911117222756991,0.9827411167512691
+17,0.014263770198030398,0.03483956679701805,0.9949125,0.988,0.9998447559375,0.999428,0.9949105278295339,0.9880239520958084
+18,0.01196307003966067,0.032166809320508034,0.995475,0.987,0.999904258125,0.999582,0.9954736420926278,0.9870388833499502
+19,0.011337511599739083,0.027693722724914552,0.9956875,0.991,0.9999116637500001,0.999648,0.9956874460930761,0.9909729187562688
+20,0.010317483875318431,0.026324373602867126,0.9964125,0.992,0.9999230290625,0.999668,0.9964117374942175,0.992
+21,0.009391166536684613,0.031228957712650298,0.99665,0.988,0.999930400625,0.999572,0.9966499162479062,0.9879759519038076
+22,0.00938724503798876,0.03080810856842436,0.9965875,0.989,0.9999409593750002,0.9996319999999999,0.9965871160505557,0.988988988988989
+23,0.00925770991302561,0.036078432086855176,0.99675,0.989,0.9999399396875,0.9995940000000001,0.9967496749674968,0.9890329012961117
+24,0.008796912119051557,0.030019028067586988,0.9971125,0.988,0.9999192143750002,0.9997039999999999,0.9971122473216406,0.9880478087649402
+25,0.009039205305592622,0.03161667108512484,0.99685,0.992,0.9999435053125,0.999608,0.9968496062007751,0.9920159680638723
+26,0.00832465733307763,0.03130640342843253,0.9968125,0.989,0.9999427559375,0.999682,0.9968128585534127,0.988988988988989
+27,0.007553178189964092,0.02917952951043844,0.9972,0.99,0.9999617928124999,0.99969,0.9972,0.9899799599198397
+28,0.007608072868280578,0.028727686047554015,0.99705,0.991,0.9999638615625001,0.999696,0.9970501474926253,0.990990990990991
+29,0.007120153656997718,0.029447013556957245,0.9973375,0.991,0.9999664453124999,0.999688,0.9973374001525057,0.990990990990991
+30,0.007157817871007137,0.02913126605004072,0.9975625,0.991,0.9999617618750001,0.999682,0.9975616160858312,0.990990990990991
+31,0.007336907451570732,0.028489415229298175,0.9971875,0.991,0.9999640784374999,0.999708,0.9971867615249003,0.991008991008991
+32,0.006501691047073109,0.03007084161043167,0.9977375,0.992,0.999968305,0.999672,0.9977373020139262,0.992
+33,0.00693205452451366,0.02848208716508816,0.9975,0.993,0.9999662856249999,0.999718,0.9974997499749975,0.993006993006993
+34,0.007461001089593629,0.028252465724945067,0.997175,0.992,0.9999625840624999,0.99971,0.997174576186428,0.992
+35,0.00710381545725686,0.029008929789066315,0.9974125,0.993,0.9999660171875,0.9997199999999999,0.9974120147527662,0.992992992992993
+36,0.006280751934862929,0.030127739772200585,0.9976625,0.992,0.999974466875,0.9997020000000001,0.9976623538971185,0.9919839679358717
+37,0.006127115182683338,0.03197662199282786,0.9976625,0.991,0.9999751071874999,0.99969,0.9976624123404627,0.990990990990991
+38,0.00666614195129805,0.030885459803044794,0.9975,0.992,0.9999712131249999,0.999694,0.9974996249437416,0.992
+39,0.007003229365494917,0.030323315382003783,0.9974375,0.992,0.9999688965624999,0.999702,0.9974374679683496,0.992
+40,0.006024414020474069,0.03197560523450375,0.9977875,0.995,0.9999746728125001,0.999688,0.997786586967124,0.995004995004995

training_notebook_a3.ipynb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:114b011c655469b68717044f932313045fe1581f964f013703e81d6d48adcb00
+size 21999529