End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8154362416107382
+      value: 0.8355704697986577
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,21 +32,20 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6445
-- Accuracy: 0.8154
+- Loss: 0.5500
+- Accuracy: 0.8356
 
 ## Model description
 
-Predict Expansion Grade - Gardner Score from an embryo image
-
+More information needed
 
 ## Intended uses & limitations
 
-More information will be provided
+More information needed
 
 ## Training and evaluation data
 
-More information will be provided
+More information needed
 
 ## Training procedure
 
@@ -62,22 +61,27 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 10
+- num_epochs: 15
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.6002        | 0.97  | 14   | 1.4558          | 0.5024   |
-| 1.4093        | 2.0   | 29   | 1.2320          | 0.5024   |
-| 1.1068        | 2.97  | 43   | 1.0740          | 0.5951   |
-| 0.9988        | 4.0   | 58   | 0.9967          | 0.6049   |
-| 0.9099        | 4.97  | 72   | 0.9248          | 0.6      |
-| 0.8674        | 6.0   | 87   | 0.8766          | 0.6780   |
-| 0.8638        | 6.97  | 101  | 0.8656          | 0.6732   |
-| 0.833         | 8.0   | 116  | 0.8395          | 0.6732   |
-| 0.8211        | 8.97  | 130  | 0.8204          | 0.6927   |
-| 0.8236        | 9.66  | 140  | 0.8204          | 0.6780   |
+| 1.6043        | 0.97  | 14   | 1.5288          | 0.5415   |
+| 1.4967        | 2.0   | 29   | 1.1719          | 0.5415   |
+| 1.1276        | 2.97  | 43   | 1.0525          | 0.5463   |
+| 1.0796        | 4.0   | 58   | 0.9086          | 0.6537   |
+| 0.9387        | 4.97  | 72   | 0.8500          | 0.6439   |
+| 0.9232        | 6.0   | 87   | 0.8190          | 0.6732   |
+| 0.8456        | 6.97  | 101  | 0.8042          | 0.6878   |
+| 0.8348        | 8.0   | 116  | 0.7770          | 0.6927   |
+| 0.8057        | 8.97  | 130  | 0.7457          | 0.7073   |
+| 0.8033        | 10.0  | 145  | 0.7353          | 0.7024   |
+| 0.7822        | 10.97 | 159  | 0.7166          | 0.7122   |
+| 0.7594        | 12.0  | 174  | 0.7188          | 0.7171   |
+| 0.7777        | 12.97 | 188  | 0.7086          | 0.7171   |
+| 0.7445        | 14.0  | 203  | 0.7139          | 0.6878   |
+| 0.7513        | 14.48 | 210  | 0.7139          | 0.6878   |
 
 
 ### Framework versions

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 9.66,
-    "eval_accuracy": 0.8154362416107382,
-    "eval_loss": 0.6445475816726685,
-    "eval_runtime": 10.3295,
-    "eval_samples_per_second": 28.849,
-    "eval_steps_per_second": 0.968,
-    "train_loss": 1.0145302023206437,
-    "train_runtime": 896.6945,
-    "train_samples_per_second": 20.509,
-    "train_steps_per_second": 0.156
+    "epoch": 14.48,
+    "eval_accuracy": 0.8355704697986577,
+    "eval_loss": 0.5500471591949463,
+    "eval_runtime": 5.0985,
+    "eval_samples_per_second": 58.448,
+    "eval_steps_per_second": 1.961,
+    "train_loss": 0.9456698463076637,
+    "train_runtime": 1102.737,
+    "train_samples_per_second": 25.015,
+    "train_steps_per_second": 0.19
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.66,
-    "eval_accuracy": 0.8154362416107382,
-    "eval_loss": 0.6445475816726685,
-    "eval_runtime": 10.3295,
-    "eval_samples_per_second": 28.849,
-    "eval_steps_per_second": 0.968
+    "epoch": 14.48,
+    "eval_accuracy": 0.8355704697986577,
+    "eval_loss": 0.5500471591949463,
+    "eval_runtime": 5.0985,
+    "eval_samples_per_second": 58.448,
+    "eval_steps_per_second": 1.961
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8780f92caaadaf679a8727782284d64e7aa4f2cb477acc21b6624ba44457c5be
+oid sha256:0ba5d11712824f6d4f6f330378764fd34aae7206d7f052e681a25944638ca39e
 size 110359372

runs/Dec27_20-56-20_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703707007.MacBook-Pro-de-Max-2.local.31343.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2310502586d42fb6eca3f41f5d1165c1cf835316b0b58c9e620aa3b3358ba1a7
+size 8608

runs/Dec27_21-09-51_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703707795.MacBook-Pro-de-Max-2.local.31343.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ac2ebe43936c7965468a39b4d27f7f686b001c9898c6e4072e421b9a4e811d6
+size 37023

runs/Dec27_21-09-51_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703719586.MacBook-Pro-de-Max-2.local.31343.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f24a62940b383067cccbd28d7f5ca0dc7266a93ef8b282cab285fca896e7984a
+size 734

runs/Dec28_00-27-59_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703719713.MacBook-Pro-de-Max-2.local.31343.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f84777390ee9fea3752cbbbdd0693698a22640b4c752e388932c01714ee1cb37
+size 5649

runs/Dec28_00-27-59_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703719811.MacBook-Pro-de-Max-2.local.31343.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6acbd6794a21eeb8a1575534f1312b0c1a0e2c943f0c6d8544908701380001c9
+size 722

runs/Dec28_00-31-35_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703719905.MacBook-Pro-de-Max-2.local.31343.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d561530c489cf72446ac0cf1c241be23dfc6a96eab829500caeadcb8298e589
+size 13243

runs/Dec28_00-31-35_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703721019.MacBook-Pro-de-Max-2.local.31343.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c2d78ab3bbe5949eb1507ea7576d7988c7963f41c07a875debbb2d0f2a01516
+size 734

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 9.66,
-    "train_loss": 1.0145302023206437,
-    "train_runtime": 896.6945,
-    "train_samples_per_second": 20.509,
-    "train_steps_per_second": 0.156
+    "epoch": 14.48,
+    "train_loss": 0.9456698463076637,
+    "train_runtime": 1102.737,
+    "train_samples_per_second": 25.015,
+    "train_steps_per_second": 0.19
 }

trainer_state.json

@@ -1,203 +1,290 @@
 {
-  "best_metric": 0.6926829268292682,
-  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-exp-max/checkpoint-130",
-  "epoch": 9.655172413793103,
+  "best_metric": 0.7170731707317073,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-exp-max/checkpoint-174",
+  "epoch": 14.482758620689655,
   "eval_steps": 500,
-  "global_step": 140,
+  "global_step": 210,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.69,
-      "learning_rate": 3.571428571428572e-05,
-      "loss": 1.6002,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 1.6043,
       "step": 10
     },
     {
       "epoch": 0.97,
-      "eval_accuracy": 0.5024390243902439,
-      "eval_loss": 1.4557533264160156,
-      "eval_runtime": 10.833,
-      "eval_samples_per_second": 18.924,
-      "eval_steps_per_second": 0.646,
+      "eval_accuracy": 0.5414634146341464,
+      "eval_loss": 1.5287535190582275,
+      "eval_runtime": 3.5003,
+      "eval_samples_per_second": 58.567,
+      "eval_steps_per_second": 2.0,
       "step": 14
     },
     {
       "epoch": 1.38,
       "learning_rate": 4.761904761904762e-05,
-      "loss": 1.4093,
+      "loss": 1.4967,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5024390243902439,
-      "eval_loss": 1.2319719791412354,
-      "eval_runtime": 4.0581,
-      "eval_samples_per_second": 50.516,
-      "eval_steps_per_second": 1.725,
+      "eval_accuracy": 0.5414634146341464,
+      "eval_loss": 1.1718776226043701,
+      "eval_runtime": 3.523,
+      "eval_samples_per_second": 58.189,
+      "eval_steps_per_second": 1.987,
       "step": 29
     },
     {
       "epoch": 2.07,
-      "learning_rate": 4.3650793650793655e-05,
-      "loss": 1.232,
+      "learning_rate": 4.761904761904762e-05,
+      "loss": 1.2615,
       "step": 30
     },
     {
       "epoch": 2.76,
-      "learning_rate": 3.968253968253968e-05,
-      "loss": 1.1068,
+      "learning_rate": 4.4973544973544974e-05,
+      "loss": 1.1276,
       "step": 40
     },
     {
       "epoch": 2.97,
-      "eval_accuracy": 0.5951219512195122,
-      "eval_loss": 1.0739854574203491,
-      "eval_runtime": 4.0516,
-      "eval_samples_per_second": 50.598,
-      "eval_steps_per_second": 1.728,
+      "eval_accuracy": 0.5463414634146342,
+      "eval_loss": 1.052482008934021,
+      "eval_runtime": 3.482,
+      "eval_samples_per_second": 58.874,
+      "eval_steps_per_second": 2.01,
       "step": 43
     },
     {
       "epoch": 3.45,
-      "learning_rate": 3.571428571428572e-05,
-      "loss": 0.9988,
+      "learning_rate": 4.232804232804233e-05,
+      "loss": 1.0796,
       "step": 50
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6048780487804878,
-      "eval_loss": 0.9966912865638733,
-      "eval_runtime": 3.5184,
-      "eval_samples_per_second": 58.266,
-      "eval_steps_per_second": 1.99,
+      "eval_accuracy": 0.6536585365853659,
+      "eval_loss": 0.9085918664932251,
+      "eval_runtime": 3.5324,
+      "eval_samples_per_second": 58.034,
+      "eval_steps_per_second": 1.982,
       "step": 58
     },
     {
       "epoch": 4.14,
-      "learning_rate": 3.1746031746031745e-05,
-      "loss": 1.0171,
+      "learning_rate": 3.968253968253968e-05,
+      "loss": 1.0019,
       "step": 60
     },
     {
       "epoch": 4.83,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.9099,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.9387,
       "step": 70
     },
     {
       "epoch": 4.97,
-      "eval_accuracy": 0.6,
-      "eval_loss": 0.9247760772705078,
-      "eval_runtime": 3.5711,
-      "eval_samples_per_second": 57.405,
-      "eval_steps_per_second": 1.96,
+      "eval_accuracy": 0.6439024390243903,
+      "eval_loss": 0.8500024676322937,
+      "eval_runtime": 3.5195,
+      "eval_samples_per_second": 58.246,
+      "eval_steps_per_second": 1.989,
       "step": 72
     },
     {
       "epoch": 5.52,
-      "learning_rate": 2.380952380952381e-05,
-      "loss": 0.8674,
+      "learning_rate": 3.439153439153439e-05,
+      "loss": 0.9232,
       "step": 80
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6780487804878049,
-      "eval_loss": 0.8766470551490784,
-      "eval_runtime": 3.5697,
-      "eval_samples_per_second": 57.428,
-      "eval_steps_per_second": 1.961,
+      "eval_accuracy": 0.6731707317073171,
+      "eval_loss": 0.8189888000488281,
+      "eval_runtime": 3.532,
+      "eval_samples_per_second": 58.041,
+      "eval_steps_per_second": 1.982,
       "step": 87
     },
     {
       "epoch": 6.21,
-      "learning_rate": 1.984126984126984e-05,
-      "loss": 0.8807,
+      "learning_rate": 3.1746031746031745e-05,
+      "loss": 0.9018,
       "step": 90
     },
     {
       "epoch": 6.9,
-      "learning_rate": 1.5873015873015872e-05,
-      "loss": 0.8638,
+      "learning_rate": 2.91005291005291e-05,
+      "loss": 0.8456,
       "step": 100
     },
     {
       "epoch": 6.97,
-      "eval_accuracy": 0.6731707317073171,
-      "eval_loss": 0.865567684173584,
-      "eval_runtime": 3.828,
-      "eval_samples_per_second": 53.553,
-      "eval_steps_per_second": 1.829,
+      "eval_accuracy": 0.6878048780487804,
+      "eval_loss": 0.8041830062866211,
+      "eval_runtime": 3.4986,
+      "eval_samples_per_second": 58.594,
+      "eval_steps_per_second": 2.001,
       "step": 101
     },
     {
       "epoch": 7.59,
-      "learning_rate": 1.1904761904761905e-05,
-      "loss": 0.833,
+      "learning_rate": 2.6455026455026456e-05,
+      "loss": 0.8348,
       "step": 110
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.6731707317073171,
-      "eval_loss": 0.8395170569419861,
-      "eval_runtime": 3.5828,
-      "eval_samples_per_second": 57.218,
-      "eval_steps_per_second": 1.954,
+      "eval_accuracy": 0.6926829268292682,
+      "eval_loss": 0.7769930362701416,
+      "eval_runtime": 3.5012,
+      "eval_samples_per_second": 58.552,
+      "eval_steps_per_second": 1.999,
       "step": 116
     },
     {
       "epoch": 8.28,
-      "learning_rate": 7.936507936507936e-06,
-      "loss": 0.8396,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 0.8287,
       "step": 120
     },
     {
       "epoch": 8.97,
-      "learning_rate": 3.968253968253968e-06,
-      "loss": 0.8211,
+      "learning_rate": 2.1164021164021164e-05,
+      "loss": 0.8057,
       "step": 130
     },
     {
       "epoch": 8.97,
-      "eval_accuracy": 0.6926829268292682,
-      "eval_loss": 0.8203723430633545,
-      "eval_runtime": 3.633,
-      "eval_samples_per_second": 56.427,
-      "eval_steps_per_second": 1.927,
+      "eval_accuracy": 0.7073170731707317,
+      "eval_loss": 0.7456830143928528,
+      "eval_runtime": 3.4827,
+      "eval_samples_per_second": 58.862,
+      "eval_steps_per_second": 2.01,
       "step": 130
     },
     {
       "epoch": 9.66,
-      "learning_rate": 0.0,
-      "loss": 0.8236,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.8033,
       "step": 140
     },
     {
-      "epoch": 9.66,
-      "eval_accuracy": 0.6780487804878049,
-      "eval_loss": 0.8203750848770142,
-      "eval_runtime": 3.5738,
-      "eval_samples_per_second": 57.361,
-      "eval_steps_per_second": 1.959,
-      "step": 140
+      "epoch": 10.0,
+      "eval_accuracy": 0.7024390243902439,
+      "eval_loss": 0.7352668642997742,
+      "eval_runtime": 3.5384,
+      "eval_samples_per_second": 57.935,
+      "eval_steps_per_second": 1.978,
+      "step": 145
     },
     {
-      "epoch": 9.66,
-      "step": 140,
-      "total_flos": 5.78085766574506e+17,
-      "train_loss": 1.0145302023206437,
-      "train_runtime": 896.6945,
-      "train_samples_per_second": 20.509,
-      "train_steps_per_second": 0.156
+      "epoch": 10.34,
+      "learning_rate": 1.5873015873015872e-05,
+      "loss": 0.7822,
+      "step": 150
+    },
+    {
+      "epoch": 10.97,
+      "eval_accuracy": 0.7121951219512195,
+      "eval_loss": 0.7165755033493042,
+      "eval_runtime": 3.4957,
+      "eval_samples_per_second": 58.643,
+      "eval_steps_per_second": 2.002,
+      "step": 159
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 1.3227513227513228e-05,
+      "loss": 0.8041,
+      "step": 160
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 1.0582010582010582e-05,
+      "loss": 0.7594,
+      "step": 170
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7170731707317073,
+      "eval_loss": 0.718829333782196,
+      "eval_runtime": 3.5638,
+      "eval_samples_per_second": 57.522,
+      "eval_steps_per_second": 1.964,
+      "step": 174
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 7.936507936507936e-06,
+      "loss": 0.7777,
+      "step": 180
+    },
+    {
+      "epoch": 12.97,
+      "eval_accuracy": 0.7170731707317073,
+      "eval_loss": 0.7086274027824402,
+      "eval_runtime": 3.4886,
+      "eval_samples_per_second": 58.762,
+      "eval_steps_per_second": 2.007,
+      "step": 188
+    },
+    {
+      "epoch": 13.1,
+      "learning_rate": 5.291005291005291e-06,
+      "loss": 0.7863,
+      "step": 190
+    },
+    {
+      "epoch": 13.79,
+      "learning_rate": 2.6455026455026455e-06,
+      "loss": 0.7445,
+      "step": 200
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.6878048780487804,
+      "eval_loss": 0.7139347791671753,
+      "eval_runtime": 3.4798,
+      "eval_samples_per_second": 58.912,
+      "eval_steps_per_second": 2.012,
+      "step": 203
+    },
+    {
+      "epoch": 14.48,
+      "learning_rate": 0.0,
+      "loss": 0.7513,
+      "step": 210
+    },
+    {
+      "epoch": 14.48,
+      "eval_accuracy": 0.6878048780487804,
+      "eval_loss": 0.7138883471488953,
+      "eval_runtime": 3.48,
+      "eval_samples_per_second": 58.908,
+      "eval_steps_per_second": 2.012,
+      "step": 210
+    },
+    {
+      "epoch": 14.48,
+      "step": 210,
+      "total_flos": 8.6685208493713e+17,
+      "train_loss": 0.9456698463076637,
+      "train_runtime": 1102.737,
+      "train_samples_per_second": 25.015,
+      "train_steps_per_second": 0.19
     }
   ],
   "logging_steps": 10,
-  "max_steps": 140,
+  "max_steps": 210,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 15,
   "save_steps": 500,
-  "total_flos": 5.78085766574506e+17,
+  "total_flos": 8.6685208493713e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1577050422d312cadf11da4407c557f68886a726d97b4971bf7def3ccd26394
+oid sha256:53217e0c5182d6e5b40d2b23665f7781a2c9b4eacf79515ea542107971a6ea22
 size 4792