End of training

Browse files

Files changed (11) hide show

README.md +108 -107
all_results.json +12 -11
config.json +71 -58
eval_results.json +7 -7
model.safetensors +1 -1
preprocessor_config.json +22 -22
runs/Oct13_15-01-08_9b69f8f7fe92/events.out.tfevents.1728831686.9b69f8f7fe92.1642.0 +3 -0
runs/Oct13_15-01-08_9b69f8f7fe92/events.out.tfevents.1728832673.9b69f8f7fe92.1642.1 +3 -0
train_results.json +7 -6
trainer_state.json +368 -348
training_args.bin +2 -2

README.md CHANGED Viewed

@@ -1,107 +1,108 @@
----
-license: apache-2.0
-base_model: microsoft/swinv2-tiny-patch4-window8-256
-tags:
-- generated_from_trainer
-datasets:
-- imagefolder
-metrics:
-- accuracy
-model-index:
-- name: swinv2-tiny-patch4-window8-256-Ocular-Toxoplasmosis
-  results:
-  - task:
-      name: Image Classification
-      type: image-classification
-    dataset:
-      name: imagefolder
-      type: imagefolder
-      config: default
-      split: validation
-      args: default
-    metrics:
-    - name: Accuracy
-      type: accuracy
-      value: 0.08064516129032258
----
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-# swinv2-tiny-patch4-window8-256-Ocular-Toxoplasmosis
-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: 8.8834
-- Accuracy: 0.0806
-## Model description
-More information needed
-## Intended uses & limitations
-More information needed
-## Training and evaluation data
-More information needed
-## Training procedure
-### Training hyperparameters
-The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 32
-- eval_batch_size: 32
-- seed: 42
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 128
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 40
-### Training results
-| Training Loss | Epoch | Step | Validation Loss | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 0.73  | 2    | 8.8834          | 0.0806   |
-| No log        | 1.82  | 5    | 8.8522          | 0.0806   |
-| No log        | 2.91  | 8    | 8.7000          | 0.0806   |
-| 8.7803        | 4.0   | 11   | 8.2692          | 0.0806   |
-| 8.7803        | 4.73  | 13   | 7.8836          | 0.0806   |
-| 8.7803        | 5.82  | 16   | 7.3279          | 0.0806   |
-| 8.7803        | 6.91  | 19   | 6.7700          | 0.0806   |
-| 7.5847        | 8.0   | 22   | 6.1880          | 0.0806   |
-| 7.5847        | 8.73  | 24   | 5.7783          | 0.0806   |
-| 7.5847        | 9.82  | 27   | 5.2113          | 0.0806   |
-| 5.7442        | 10.91 | 30   | 4.7163          | 0.0806   |
-| 5.7442        | 12.0  | 33   | 4.2648          | 0.0806   |
-| 5.7442        | 12.73 | 35   | 3.9892          | 0.0806   |
-| 5.7442        | 13.82 | 38   | 3.6134          | 0.0806   |
-| 4.1747        | 14.91 | 41   | 3.2828          | 0.0806   |
-| 4.1747        | 16.0  | 44   | 2.9957          | 0.0806   |
-| 4.1747        | 16.73 | 46   | 2.8259          | 0.0806   |
-| 4.1747        | 17.82 | 49   | 2.5988          | 0.0806   |
-| 3.0458        | 18.91 | 52   | 2.4004          | 0.0806   |
-| 3.0458        | 20.0  | 55   | 2.2272          | 0.0806   |
-| 3.0458        | 20.73 | 57   | 2.1254          | 0.0806   |
-| 2.3301        | 21.82 | 60   | 1.9937          | 0.0806   |
-| 2.3301        | 22.91 | 63   | 1.8860          | 0.0806   |
-| 2.3301        | 24.0  | 66   | 1.8005          | 0.0806   |
-| 2.3301        | 24.73 | 68   | 1.7551          | 0.0806   |
-| 1.9107        | 25.82 | 71   | 1.7021          | 0.0806   |
-| 1.9107        | 26.91 | 74   | 1.6654          | 0.0806   |
-| 1.9107        | 28.0  | 77   | 1.6434          | 0.0806   |
-| 1.9107        | 28.73 | 79   | 1.6362          | 0.0806   |
-| 1.7061        | 29.09 | 80   | 1.6348          | 0.0806   |
-### Framework versions
-- Transformers 4.36.2
-- Pytorch 2.1.2+cu118
-- Datasets 2.16.1
-- Tokenizers 0.15.0

+---
+library_name: transformers
+license: apache-2.0
+base_model: microsoft/swinv2-tiny-patch4-window8-256
+tags:
+- generated_from_trainer
+datasets:
+- imagefolder
+metrics:
+- accuracy
+model-index:
+- name: swinv2-tiny-patch4-window8-256-Ocular-Toxoplasmosis
+  results:
+  - task:
+      name: Image Classification
+      type: image-classification
+    dataset:
+      name: imagefolder
+      type: imagefolder
+      config: default
+      split: validation
+      args: default
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.8387096774193549
+---
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+# swinv2-tiny-patch4-window8-256-Ocular-Toxoplasmosis
+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: 0.5167
+- Accuracy: 0.8387
+## Model description
+More information needed
+## Intended uses & limitations
+More information needed
+## Training and evaluation data
+More information needed
+## Training procedure
+### Training hyperparameters
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 128
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 40
+### Training results
+| Training Loss | Epoch   | Step | Validation Loss | Accuracy |
+|:-------------:|:-------:|:----:|:---------------:|:--------:|
+| No log        | 0.7273  | 2    | 1.4057          | 0.2419   |
+| No log        | 1.8182  | 5    | 1.2100          | 0.4677   |
+| No log        | 2.9091  | 8    | 1.1808          | 0.4516   |
+| 1.3062        | 4.0     | 11   | 1.0975          | 0.5968   |
+| 1.3062        | 4.7273  | 13   | 1.0542          | 0.6613   |
+| 1.3062        | 5.8182  | 16   | 0.9857          | 0.6613   |
+| 1.3062        | 6.9091  | 19   | 0.9176          | 0.6774   |
+| 1.0003        | 8.0     | 22   | 0.8761          | 0.6774   |
+| 1.0003        | 8.7273  | 24   | 0.8540          | 0.6774   |
+| 1.0003        | 9.8182  | 27   | 0.7777          | 0.6613   |
+| 0.8096        | 10.9091 | 30   | 0.7498          | 0.6613   |
+| 0.8096        | 12.0    | 33   | 0.7569          | 0.6613   |
+| 0.8096        | 12.7273 | 35   | 0.7422          | 0.6774   |
+| 0.8096        | 13.8182 | 38   | 0.7278          | 0.7097   |
+| 0.6556        | 14.9091 | 41   | 0.6877          | 0.7258   |
+| 0.6556        | 16.0    | 44   | 0.6433          | 0.7258   |
+| 0.6556        | 16.7273 | 46   | 0.6324          | 0.7419   |
+| 0.6556        | 17.8182 | 49   | 0.6390          | 0.7419   |
+| 0.5725        | 18.9091 | 52   | 0.6504          | 0.7742   |
+| 0.5725        | 20.0    | 55   | 0.6145          | 0.7581   |
+| 0.5725        | 20.7273 | 57   | 0.5824          | 0.7903   |
+| 0.5057        | 21.8182 | 60   | 0.5476          | 0.8226   |
+| 0.5057        | 22.9091 | 63   | 0.5413          | 0.8226   |
+| 0.5057        | 24.0    | 66   | 0.5335          | 0.8226   |
+| 0.5057        | 24.7273 | 68   | 0.5302          | 0.8226   |
+| 0.4945        | 25.8182 | 71   | 0.5231          | 0.8226   |
+| 0.4945        | 26.9091 | 74   | 0.5167          | 0.8387   |
+| 0.4945        | 28.0    | 77   | 0.5132          | 0.8387   |
+| 0.4945        | 28.7273 | 79   | 0.5131          | 0.8387   |
+| 0.4883        | 29.0909 | 80   | 0.5131          | 0.8387   |
+### Framework versions
+- Transformers 4.44.2
+- Pytorch 2.4.1+cu121
+- Datasets 3.0.1
+- Tokenizers 0.19.1

all_results.json CHANGED Viewed

@@ -1,12 +1,13 @@
-{
-    "epoch": 29.09,
-    "eval_accuracy": 0.08064516129032258,
-    "eval_loss": 8.883430480957031,
-    "eval_runtime": 2.5622,
-    "eval_samples_per_second": 24.198,
-    "eval_steps_per_second": 0.781,
-    "train_loss": 4.409568953514099,
-    "train_runtime": 541.1993,
-    "train_samples_per_second": 25.868,
-    "train_steps_per_second": 0.148
 }

+{
+    "epoch": 29.09090909090909,
+    "eval_accuracy": 0.8387096774193549,
+    "eval_loss": 0.5166797637939453,
+    "eval_runtime": 2.8799,
+    "eval_samples_per_second": 21.529,
+    "eval_steps_per_second": 0.694,
+    "total_flos": 3.312830060612813e+17,
+    "train_loss": 0.7290899336338044,
+    "train_runtime": 714.1327,
+    "train_samples_per_second": 19.604,
+    "train_steps_per_second": 0.112
 }

config.json CHANGED Viewed

@@ -1,58 +1,71 @@
-{
-  "_name_or_path": "microsoft/swinv2-tiny-patch4-window8-256",
-  "architectures": [
-    "Swinv2ForImageClassification"
-  ],
-  "attention_probs_dropout_prob": 0.0,
-  "depths": [
-    2,
-    2,
-    6,
-    2
-  ],
-  "drop_path_rate": 0.1,
-  "embed_dim": 96,
-  "encoder_stride": 32,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.0,
-  "hidden_size": 768,
-  "id2label": {
-    "0": "active",
-    "1": "active-inactive",
-    "2": "healthy",
-    "3": "inactive"
-  },
-  "image_size": 256,
-  "initializer_range": 0.02,
-  "label2id": {
-    "active": 0,
-    "active-inactive": 1,
-    "healthy": 2,
-    "inactive": 3
-  },
-  "layer_norm_eps": 1e-05,
-  "mlp_ratio": 4.0,
-  "model_type": "swinv2",
-  "num_channels": 3,
-  "num_heads": [
-    3,
-    6,
-    12,
-    24
-  ],
-  "num_layers": 4,
-  "patch_size": 4,
-  "path_norm": true,
-  "pretrained_window_sizes": [
-    0,
-    0,
-    0,
-    0
-  ],
-  "problem_type": "single_label_classification",
-  "qkv_bias": true,
-  "torch_dtype": "float32",
-  "transformers_version": "4.36.2",
-  "use_absolute_embeddings": false,
-  "window_size": 8
-}

+{
+  "_name_or_path": "microsoft/swinv2-tiny-patch4-window8-256",
+  "architectures": [
+    "Swinv2ForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "depths": [
+    2,
+    2,
+    6,
+    2
+  ],
+  "drop_path_rate": 0.1,
+  "embed_dim": 96,
+  "encoder_stride": 32,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "active",
+    "1": "active-inactive",
+    "2": "healthy",
+    "3": "inactive"
+  },
+  "image_size": 256,
+  "initializer_range": 0.02,
+  "label2id": {
+    "active": 0,
+    "active-inactive": 1,
+    "healthy": 2,
+    "inactive": 3
+  },
+  "layer_norm_eps": 1e-05,
+  "mlp_ratio": 4.0,
+  "model_type": "swinv2",
+  "num_channels": 3,
+  "num_heads": [
+    3,
+    6,
+    12,
+    24
+  ],
+  "num_layers": 4,
+  "out_features": [
+    "stage4"
+  ],
+  "out_indices": [
+    4
+  ],
+  "patch_size": 4,
+  "path_norm": true,
+  "pretrained_window_sizes": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "stage_names": [
+    "stem",
+    "stage1",
+    "stage2",
+    "stage3",
+    "stage4"
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2",
+  "use_absolute_embeddings": false,
+  "window_size": 8
+}

eval_results.json CHANGED Viewed

@@ -1,8 +1,8 @@
-{
-    "epoch": 29.09,
-    "eval_accuracy": 0.08064516129032258,
-    "eval_loss": 8.883430480957031,
-    "eval_runtime": 2.5622,
-    "eval_samples_per_second": 24.198,
-    "eval_steps_per_second": 0.781
 }

+{
+    "epoch": 29.09090909090909,
+    "eval_accuracy": 0.8387096774193549,
+    "eval_loss": 0.5166797637939453,
+    "eval_runtime": 2.8799,
+    "eval_samples_per_second": 21.529,
+    "eval_steps_per_second": 0.694
 }

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2046d4398c99900e190202d9adf3e1f7972aa1a911326e467b32625d0cee42f5
 size 110356296

 version https://git-lfs.github.com/spec/v1
+oid sha256:9088d5cdfd4e0544dc4cb2872f795548dc215c7272f7a2578464c3f1ac9e198c
 size 110356296

preprocessor_config.json CHANGED Viewed

@@ -1,22 +1,22 @@
-{
-  "do_normalize": true,
-  "do_rescale": true,
-  "do_resize": true,
-  "image_mean": [
-    0.485,
-    0.456,
-    0.406
-  ],
-  "image_processor_type": "ViTImageProcessor",
-  "image_std": [
-    0.229,
-    0.224,
-    0.225
-  ],
-  "resample": 3,
-  "rescale_factor": 0.00392156862745098,
-  "size": {
-    "height": 256,
-    "width": 256
-  }
-}

+{
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 256,
+    "width": 256
+  }
+}

runs/Oct13_15-01-08_9b69f8f7fe92/events.out.tfevents.1728831686.9b69f8f7fe92.1642.0 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:67e20f350dc036eac5e18af66e5effe677d91038def2cbec8e066bc3a47c286c
+size 17079

runs/Oct13_15-01-08_9b69f8f7fe92/events.out.tfevents.1728832673.9b69f8f7fe92.1642.1 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b056a3326bc87b591abc5b6b3d29054361e4ede4d40769bbc30de720bc87c957
+size 405

train_results.json CHANGED Viewed

@@ -1,7 +1,8 @@
-{
-    "epoch": 29.09,
-    "train_loss": 4.409568953514099,
-    "train_runtime": 541.1993,
-    "train_samples_per_second": 25.868,
-    "train_steps_per_second": 0.148
 }

+{
+    "epoch": 29.09090909090909,
+    "total_flos": 3.312830060612813e+17,
+    "train_loss": 0.7290899336338044,
+    "train_runtime": 714.1327,
+    "train_samples_per_second": 19.604,
+    "train_steps_per_second": 0.112
 }

trainer_state.json CHANGED Viewed

@@ -1,348 +1,368 @@
-{
-  "best_metric": 0.08064516129032258,
-  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-Ocular-Toxoplasmosis\\checkpoint-2",
-  "epoch": 29.09090909090909,
-  "eval_steps": 500,
-  "global_step": 80,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 8.883430480957031,
-      "eval_runtime": 2.2637,
-      "eval_samples_per_second": 27.389,
-      "eval_steps_per_second": 0.884,
-      "step": 2
-    },
-    {
-      "epoch": 1.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 8.852208137512207,
-      "eval_runtime": 2.3018,
-      "eval_samples_per_second": 26.935,
-      "eval_steps_per_second": 0.869,
-      "step": 5
-    },
-    {
-      "epoch": 2.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 8.700010299682617,
-      "eval_runtime": 2.5761,
-      "eval_samples_per_second": 24.068,
-      "eval_steps_per_second": 0.776,
-      "step": 8
-    },
-    {
-      "epoch": 3.64,
-      "learning_rate": 4.8611111111111115e-05,
-      "loss": 8.7803,
-      "step": 10
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 8.269183158874512,
-      "eval_runtime": 2.5329,
-      "eval_samples_per_second": 24.478,
-      "eval_steps_per_second": 0.79,
-      "step": 11
-    },
-    {
-      "epoch": 4.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 7.88364839553833,
-      "eval_runtime": 2.4337,
-      "eval_samples_per_second": 25.475,
-      "eval_steps_per_second": 0.822,
-      "step": 13
-    },
-    {
-      "epoch": 5.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 7.327876091003418,
-      "eval_runtime": 2.5074,
-      "eval_samples_per_second": 24.727,
-      "eval_steps_per_second": 0.798,
-      "step": 16
-    },
-    {
-      "epoch": 6.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 6.769954204559326,
-      "eval_runtime": 2.6471,
-      "eval_samples_per_second": 23.422,
-      "eval_steps_per_second": 0.756,
-      "step": 19
-    },
-    {
-      "epoch": 7.27,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 7.5847,
-      "step": 20
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 6.1880202293396,
-      "eval_runtime": 2.602,
-      "eval_samples_per_second": 23.828,
-      "eval_steps_per_second": 0.769,
-      "step": 22
-    },
-    {
-      "epoch": 8.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 5.778294563293457,
-      "eval_runtime": 2.4341,
-      "eval_samples_per_second": 25.471,
-      "eval_steps_per_second": 0.822,
-      "step": 24
-    },
-    {
-      "epoch": 9.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 5.21131706237793,
-      "eval_runtime": 2.3164,
-      "eval_samples_per_second": 26.766,
-      "eval_steps_per_second": 0.863,
-      "step": 27
-    },
-    {
-      "epoch": 10.91,
-      "learning_rate": 3.472222222222222e-05,
-      "loss": 5.7442,
-      "step": 30
-    },
-    {
-      "epoch": 10.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 4.716261386871338,
-      "eval_runtime": 2.4233,
-      "eval_samples_per_second": 25.585,
-      "eval_steps_per_second": 0.825,
-      "step": 30
-    },
-    {
-      "epoch": 12.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 4.264786720275879,
-      "eval_runtime": 2.513,
-      "eval_samples_per_second": 24.671,
-      "eval_steps_per_second": 0.796,
-      "step": 33
-    },
-    {
-      "epoch": 12.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 3.989229202270508,
-      "eval_runtime": 2.4651,
-      "eval_samples_per_second": 25.151,
-      "eval_steps_per_second": 0.811,
-      "step": 35
-    },
-    {
-      "epoch": 13.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 3.6134493350982666,
-      "eval_runtime": 2.6037,
-      "eval_samples_per_second": 23.812,
-      "eval_steps_per_second": 0.768,
-      "step": 38
-    },
-    {
-      "epoch": 14.55,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 4.1747,
-      "step": 40
-    },
-    {
-      "epoch": 14.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 3.2827646732330322,
-      "eval_runtime": 2.687,
-      "eval_samples_per_second": 23.074,
-      "eval_steps_per_second": 0.744,
-      "step": 41
-    },
-    {
-      "epoch": 16.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 2.9957385063171387,
-      "eval_runtime": 2.4174,
-      "eval_samples_per_second": 25.647,
-      "eval_steps_per_second": 0.827,
-      "step": 44
-    },
-    {
-      "epoch": 16.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 2.825892686843872,
-      "eval_runtime": 2.3083,
-      "eval_samples_per_second": 26.86,
-      "eval_steps_per_second": 0.866,
-      "step": 46
-    },
-    {
-      "epoch": 17.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 2.5987932682037354,
-      "eval_runtime": 2.4694,
-      "eval_samples_per_second": 25.107,
-      "eval_steps_per_second": 0.81,
-      "step": 49
-    },
-    {
-      "epoch": 18.18,
-      "learning_rate": 2.0833333333333336e-05,
-      "loss": 3.0458,
-      "step": 50
-    },
-    {
-      "epoch": 18.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 2.400411367416382,
-      "eval_runtime": 2.3426,
-      "eval_samples_per_second": 26.467,
-      "eval_steps_per_second": 0.854,
-      "step": 52
-    },
-    {
-      "epoch": 20.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 2.227222204208374,
-      "eval_runtime": 2.4914,
-      "eval_samples_per_second": 24.885,
-      "eval_steps_per_second": 0.803,
-      "step": 55
-    },
-    {
-      "epoch": 20.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 2.125420331954956,
-      "eval_runtime": 2.3746,
-      "eval_samples_per_second": 26.11,
-      "eval_steps_per_second": 0.842,
-      "step": 57
-    },
-    {
-      "epoch": 21.82,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 2.3301,
-      "step": 60
-    },
-    {
-      "epoch": 21.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.9937151670455933,
-      "eval_runtime": 2.4362,
-      "eval_samples_per_second": 25.449,
-      "eval_steps_per_second": 0.821,
-      "step": 60
-    },
-    {
-      "epoch": 22.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.885993242263794,
-      "eval_runtime": 2.4078,
-      "eval_samples_per_second": 25.749,
-      "eval_steps_per_second": 0.831,
-      "step": 63
-    },
-    {
-      "epoch": 24.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.8005385398864746,
-      "eval_runtime": 2.3561,
-      "eval_samples_per_second": 26.314,
-      "eval_steps_per_second": 0.849,
-      "step": 66
-    },
-    {
-      "epoch": 24.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.7550740242004395,
-      "eval_runtime": 2.3863,
-      "eval_samples_per_second": 25.981,
-      "eval_steps_per_second": 0.838,
-      "step": 68
-    },
-    {
-      "epoch": 25.45,
-      "learning_rate": 6.944444444444445e-06,
-      "loss": 1.9107,
-      "step": 70
-    },
-    {
-      "epoch": 25.82,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.7021311521530151,
-      "eval_runtime": 2.3225,
-      "eval_samples_per_second": 26.696,
-      "eval_steps_per_second": 0.861,
-      "step": 71
-    },
-    {
-      "epoch": 26.91,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.6653900146484375,
-      "eval_runtime": 2.59,
-      "eval_samples_per_second": 23.939,
-      "eval_steps_per_second": 0.772,
-      "step": 74
-    },
-    {
-      "epoch": 28.0,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.6433522701263428,
-      "eval_runtime": 2.5188,
-      "eval_samples_per_second": 24.615,
-      "eval_steps_per_second": 0.794,
-      "step": 77
-    },
-    {
-      "epoch": 28.73,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.6361864805221558,
-      "eval_runtime": 2.3834,
-      "eval_samples_per_second": 26.013,
-      "eval_steps_per_second": 0.839,
-      "step": 79
-    },
-    {
-      "epoch": 29.09,
-      "learning_rate": 0.0,
-      "loss": 1.7061,
-      "step": 80
-    },
-    {
-      "epoch": 29.09,
-      "eval_accuracy": 0.08064516129032258,
-      "eval_loss": 1.6347676515579224,
-      "eval_runtime": 2.4175,
-      "eval_samples_per_second": 25.646,
-      "eval_steps_per_second": 0.827,
-      "step": 80
-    },
-    {
-      "epoch": 29.09,
-      "step": 80,
-      "total_flos": 3.312830060612813e+17,
-      "train_loss": 4.409568953514099,
-      "train_runtime": 541.1993,
-      "train_samples_per_second": 25.868,
-      "train_steps_per_second": 0.148
-    }
-  ],
-  "logging_steps": 10,
-  "max_steps": 80,
-  "num_input_tokens_seen": 0,
-  "num_train_epochs": 40,
-  "save_steps": 500,
-  "total_flos": 3.312830060612813e+17,
-  "train_batch_size": 32,
-  "trial_name": null,
-  "trial_params": null
-}

+{
+  "best_metric": 0.8387096774193549,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-Ocular-Toxoplasmosis/checkpoint-74",
+  "epoch": 29.09090909090909,
+  "eval_steps": 500,
+  "global_step": 80,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.7272727272727273,
+      "eval_accuracy": 0.24193548387096775,
+      "eval_loss": 1.4057228565216064,
+      "eval_runtime": 3.9176,
+      "eval_samples_per_second": 15.826,
+      "eval_steps_per_second": 0.511,
+      "step": 2
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "eval_accuracy": 0.46774193548387094,
+      "eval_loss": 1.2099871635437012,
+      "eval_runtime": 2.8828,
+      "eval_samples_per_second": 21.507,
+      "eval_steps_per_second": 0.694,
+      "step": 5
+    },
+    {
+      "epoch": 2.909090909090909,
+      "eval_accuracy": 0.45161290322580644,
+      "eval_loss": 1.18076491355896,
+      "eval_runtime": 2.8384,
+      "eval_samples_per_second": 21.843,
+      "eval_steps_per_second": 0.705,
+      "step": 8
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 10.30179214477539,
+      "learning_rate": 4.8611111111111115e-05,
+      "loss": 1.3062,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5967741935483871,
+      "eval_loss": 1.0975382328033447,
+      "eval_runtime": 3.6135,
+      "eval_samples_per_second": 17.158,
+      "eval_steps_per_second": 0.553,
+      "step": 11
+    },
+    {
+      "epoch": 4.7272727272727275,
+      "eval_accuracy": 0.6612903225806451,
+      "eval_loss": 1.0542328357696533,
+      "eval_runtime": 3.5171,
+      "eval_samples_per_second": 17.628,
+      "eval_steps_per_second": 0.569,
+      "step": 13
+    },
+    {
+      "epoch": 5.818181818181818,
+      "eval_accuracy": 0.6612903225806451,
+      "eval_loss": 0.9857348799705505,
+      "eval_runtime": 2.887,
+      "eval_samples_per_second": 21.475,
+      "eval_steps_per_second": 0.693,
+      "step": 16
+    },
+    {
+      "epoch": 6.909090909090909,
+      "eval_accuracy": 0.6774193548387096,
+      "eval_loss": 0.9176284074783325,
+      "eval_runtime": 2.8754,
+      "eval_samples_per_second": 21.562,
+      "eval_steps_per_second": 0.696,
+      "step": 19
+    },
+    {
+      "epoch": 7.2727272727272725,
+      "grad_norm": 4.642858982086182,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 1.0003,
+      "step": 20
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6774193548387096,
+      "eval_loss": 0.8760596513748169,
+      "eval_runtime": 3.7173,
+      "eval_samples_per_second": 16.679,
+      "eval_steps_per_second": 0.538,
+      "step": 22
+    },
+    {
+      "epoch": 8.727272727272727,
+      "eval_accuracy": 0.6774193548387096,
+      "eval_loss": 0.8539677262306213,
+      "eval_runtime": 3.3041,
+      "eval_samples_per_second": 18.764,
+      "eval_steps_per_second": 0.605,
+      "step": 24
+    },
+    {
+      "epoch": 9.818181818181818,
+      "eval_accuracy": 0.6612903225806451,
+      "eval_loss": 0.7776592969894409,
+      "eval_runtime": 3.1776,
+      "eval_samples_per_second": 19.511,
+      "eval_steps_per_second": 0.629,
+      "step": 27
+    },
+    {
+      "epoch": 10.909090909090908,
+      "grad_norm": 5.499239921569824,
+      "learning_rate": 3.472222222222222e-05,
+      "loss": 0.8096,
+      "step": 30
+    },
+    {
+      "epoch": 10.909090909090908,
+      "eval_accuracy": 0.6612903225806451,
+      "eval_loss": 0.7497676014900208,
+      "eval_runtime": 3.1503,
+      "eval_samples_per_second": 19.68,
+      "eval_steps_per_second": 0.635,
+      "step": 30
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.6612903225806451,
+      "eval_loss": 0.7568932175636292,
+      "eval_runtime": 3.7657,
+      "eval_samples_per_second": 16.465,
+      "eval_steps_per_second": 0.531,
+      "step": 33
+    },
+    {
+      "epoch": 12.727272727272727,
+      "eval_accuracy": 0.6774193548387096,
+      "eval_loss": 0.7422052025794983,
+      "eval_runtime": 3.1158,
+      "eval_samples_per_second": 19.898,
+      "eval_steps_per_second": 0.642,
+      "step": 35
+    },
+    {
+      "epoch": 13.818181818181818,
+      "eval_accuracy": 0.7096774193548387,
+      "eval_loss": 0.7278109788894653,
+      "eval_runtime": 2.8488,
+      "eval_samples_per_second": 21.763,
+      "eval_steps_per_second": 0.702,
+      "step": 38
+    },
+    {
+      "epoch": 14.545454545454545,
+      "grad_norm": 8.175309181213379,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.6556,
+      "step": 40
+    },
+    {
+      "epoch": 14.909090909090908,
+      "eval_accuracy": 0.7258064516129032,
+      "eval_loss": 0.687738835811615,
+      "eval_runtime": 2.8406,
+      "eval_samples_per_second": 21.827,
+      "eval_steps_per_second": 0.704,
+      "step": 41
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.7258064516129032,
+      "eval_loss": 0.6433460116386414,
+      "eval_runtime": 4.134,
+      "eval_samples_per_second": 14.998,
+      "eval_steps_per_second": 0.484,
+      "step": 44
+    },
+    {
+      "epoch": 16.727272727272727,
+      "eval_accuracy": 0.7419354838709677,
+      "eval_loss": 0.6324245929718018,
+      "eval_runtime": 2.8555,
+      "eval_samples_per_second": 21.713,
+      "eval_steps_per_second": 0.7,
+      "step": 46
+    },
+    {
+      "epoch": 17.818181818181817,
+      "eval_accuracy": 0.7419354838709677,
+      "eval_loss": 0.6389685273170471,
+      "eval_runtime": 2.8092,
+      "eval_samples_per_second": 22.07,
+      "eval_steps_per_second": 0.712,
+      "step": 49
+    },
+    {
+      "epoch": 18.181818181818183,
+      "grad_norm": 5.849218845367432,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 0.5725,
+      "step": 50
+    },
+    {
+      "epoch": 18.90909090909091,
+      "eval_accuracy": 0.7741935483870968,
+      "eval_loss": 0.6503620743751526,
+      "eval_runtime": 2.8945,
+      "eval_samples_per_second": 21.42,
+      "eval_steps_per_second": 0.691,
+      "step": 52
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.7580645161290323,
+      "eval_loss": 0.6144644618034363,
+      "eval_runtime": 4.0673,
+      "eval_samples_per_second": 15.244,
+      "eval_steps_per_second": 0.492,
+      "step": 55
+    },
+    {
+      "epoch": 20.727272727272727,
+      "eval_accuracy": 0.7903225806451613,
+      "eval_loss": 0.5823854207992554,
+      "eval_runtime": 3.0464,
+      "eval_samples_per_second": 20.352,
+      "eval_steps_per_second": 0.657,
+      "step": 57
+    },
+    {
+      "epoch": 21.818181818181817,
+      "grad_norm": 6.505163669586182,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.5057,
+      "step": 60
+    },
+    {
+      "epoch": 21.818181818181817,
+      "eval_accuracy": 0.8225806451612904,
+      "eval_loss": 0.547602117061615,
+      "eval_runtime": 2.9115,
+      "eval_samples_per_second": 21.295,
+      "eval_steps_per_second": 0.687,
+      "step": 60
+    },
+    {
+      "epoch": 22.90909090909091,
+      "eval_accuracy": 0.8225806451612904,
+      "eval_loss": 0.5412537455558777,
+      "eval_runtime": 2.8707,
+      "eval_samples_per_second": 21.598,
+      "eval_steps_per_second": 0.697,
+      "step": 63
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.8225806451612904,
+      "eval_loss": 0.5334817171096802,
+      "eval_runtime": 3.2898,
+      "eval_samples_per_second": 18.846,
+      "eval_steps_per_second": 0.608,
+      "step": 66
+    },
+    {
+      "epoch": 24.727272727272727,
+      "eval_accuracy": 0.8225806451612904,
+      "eval_loss": 0.5301870703697205,
+      "eval_runtime": 3.8383,
+      "eval_samples_per_second": 16.153,
+      "eval_steps_per_second": 0.521,
+      "step": 68
+    },
+    {
+      "epoch": 25.454545454545453,
+      "grad_norm": 8.245793342590332,
+      "learning_rate": 6.944444444444445e-06,
+      "loss": 0.4945,
+      "step": 70
+    },
+    {
+      "epoch": 25.818181818181817,
+      "eval_accuracy": 0.8225806451612904,
+      "eval_loss": 0.5231319665908813,
+      "eval_runtime": 3.1472,
+      "eval_samples_per_second": 19.7,
+      "eval_steps_per_second": 0.635,
+      "step": 71
+    },
+    {
+      "epoch": 26.90909090909091,
+      "eval_accuracy": 0.8387096774193549,
+      "eval_loss": 0.5166797637939453,
+      "eval_runtime": 3.151,
+      "eval_samples_per_second": 19.677,
+      "eval_steps_per_second": 0.635,
+      "step": 74
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.8387096774193549,
+      "eval_loss": 0.5131666660308838,
+      "eval_runtime": 3.202,
+      "eval_samples_per_second": 19.363,
+      "eval_steps_per_second": 0.625,
+      "step": 77
+    },
+    {
+      "epoch": 28.727272727272727,
+      "eval_accuracy": 0.8387096774193549,
+      "eval_loss": 0.513070821762085,
+      "eval_runtime": 4.4164,
+      "eval_samples_per_second": 14.039,
+      "eval_steps_per_second": 0.453,
+      "step": 79
+    },
+    {
+      "epoch": 29.09090909090909,
+      "grad_norm": 9.397185325622559,
+      "learning_rate": 0.0,
+      "loss": 0.4883,
+      "step": 80
+    },
+    {
+      "epoch": 29.09090909090909,
+      "eval_accuracy": 0.8387096774193549,
+      "eval_loss": 0.5131446719169617,
+      "eval_runtime": 2.9088,
+      "eval_samples_per_second": 21.314,
+      "eval_steps_per_second": 0.688,
+      "step": 80
+    },
+    {
+      "epoch": 29.09090909090909,
+      "step": 80,
+      "total_flos": 3.312830060612813e+17,
+      "train_loss": 0.7290899336338044,
+      "train_runtime": 714.1327,
+      "train_samples_per_second": 19.604,
+      "train_steps_per_second": 0.112
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 80,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 40,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.312830060612813e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:707b23c3a5fc468561b9ad9ec6c5cb53ee88b1b9a1f9cd003dd50ee2da9987b5
-size 4792

 version https://git-lfs.github.com/spec/v1
+oid sha256:d721724455b9c9117091e3482c977b759172e83b66a5f194bff2b562307699ec
+size 5304