End of training

README.md

@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/resnet-50](https://huggingface.co/microsoft/resnet-50) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6800
+- Loss: 0.6922
 - Accuracy: 0.9310
 
 ## Model description

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 9.655172413793103,
+    "eval_accuracy": 0.9310344827586207,
+    "eval_loss": 0.6921818852424622,
+    "eval_runtime": 0.1602,
+    "eval_samples_per_second": 181.05,
+    "eval_steps_per_second": 49.945,
+    "total_flos": 2.321029594662912e+16,
+    "train_loss": 0.6887211867741176,
+    "train_runtime": 19.8811,
+    "train_samples_per_second": 56.838,
+    "train_steps_per_second": 3.521
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.655172413793103,
+    "eval_accuracy": 0.9310344827586207,
+    "eval_loss": 0.6921818852424622,
+    "eval_runtime": 0.1602,
+    "eval_samples_per_second": 181.05,
+    "eval_steps_per_second": 49.945
+}

runs/Aug12_22-57-31_EMIMDGXA100GPU1/events.out.tfevents.1723517875.EMIMDGXA100GPU1.76048.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99229167e41d39d56d3b61b3af676ede75ecb5f9afafca9b7bd1ac9ae620110c
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.655172413793103,
+    "total_flos": 2.321029594662912e+16,
+    "train_loss": 0.6887211867741176,
+    "train_runtime": 19.8811,
+    "train_samples_per_second": 56.838,
+    "train_steps_per_second": 3.521
+}

trainer_state.json

@@ -0,0 +1,181 @@
+{
+  "best_metric": 0.9310344827586207,
+  "best_model_checkpoint": "resnet-50/checkpoint-7",
+  "epoch": 9.655172413793103,
+  "eval_steps": 500,
+  "global_step": 70,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.9655172413793104,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6921818852424622,
+      "eval_runtime": 0.1674,
+      "eval_samples_per_second": 173.228,
+      "eval_steps_per_second": 47.787,
+      "step": 7
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 2.6714026927948,
+      "learning_rate": 4.761904761904762e-06,
+      "loss": 0.6927,
+      "step": 10
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6895474791526794,
+      "eval_runtime": 0.1422,
+      "eval_samples_per_second": 203.982,
+      "eval_steps_per_second": 56.271,
+      "step": 14
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 1.4340662956237793,
+      "learning_rate": 3.968253968253968e-06,
+      "loss": 0.6916,
+      "step": 20
+    },
+    {
+      "epoch": 2.896551724137931,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6877713203430176,
+      "eval_runtime": 0.1339,
+      "eval_samples_per_second": 216.634,
+      "eval_steps_per_second": 59.761,
+      "step": 21
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6853156089782715,
+      "eval_runtime": 0.15,
+      "eval_samples_per_second": 193.285,
+      "eval_steps_per_second": 53.32,
+      "step": 29
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 0.846211314201355,
+      "learning_rate": 3.1746031746031746e-06,
+      "loss": 0.6899,
+      "step": 30
+    },
+    {
+      "epoch": 4.9655172413793105,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6838722825050354,
+      "eval_runtime": 0.1539,
+      "eval_samples_per_second": 188.443,
+      "eval_steps_per_second": 51.984,
+      "step": 36
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 0.7562207579612732,
+      "learning_rate": 2.380952380952381e-06,
+      "loss": 0.6878,
+      "step": 40
+    },
+    {
+      "epoch": 5.931034482758621,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6810795068740845,
+      "eval_runtime": 0.1817,
+      "eval_samples_per_second": 159.574,
+      "eval_steps_per_second": 44.02,
+      "step": 43
+    },
+    {
+      "epoch": 6.896551724137931,
+      "grad_norm": 1.3402385711669922,
+      "learning_rate": 1.5873015873015873e-06,
+      "loss": 0.6868,
+      "step": 50
+    },
+    {
+      "epoch": 6.896551724137931,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6826210618019104,
+      "eval_runtime": 0.1474,
+      "eval_samples_per_second": 196.8,
+      "eval_steps_per_second": 54.29,
+      "step": 50
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6804185509681702,
+      "eval_runtime": 0.1782,
+      "eval_samples_per_second": 162.726,
+      "eval_steps_per_second": 44.89,
+      "step": 58
+    },
+    {
+      "epoch": 8.275862068965518,
+      "grad_norm": 1.5807445049285889,
+      "learning_rate": 7.936507936507937e-07,
+      "loss": 0.6864,
+      "step": 60
+    },
+    {
+      "epoch": 8.96551724137931,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6801003217697144,
+      "eval_runtime": 0.1994,
+      "eval_samples_per_second": 145.402,
+      "eval_steps_per_second": 40.111,
+      "step": 65
+    },
+    {
+      "epoch": 9.655172413793103,
+      "grad_norm": 0.9159764051437378,
+      "learning_rate": 0.0,
+      "loss": 0.686,
+      "step": 70
+    },
+    {
+      "epoch": 9.655172413793103,
+      "eval_accuracy": 0.9310344827586207,
+      "eval_loss": 0.6800179481506348,
+      "eval_runtime": 0.1551,
+      "eval_samples_per_second": 186.997,
+      "eval_steps_per_second": 51.585,
+      "step": 70
+    },
+    {
+      "epoch": 9.655172413793103,
+      "step": 70,
+      "total_flos": 2.321029594662912e+16,
+      "train_loss": 0.6887211867741176,
+      "train_runtime": 19.8811,
+      "train_samples_per_second": 56.838,
+      "train_steps_per_second": 3.521
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 70,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "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": 2.321029594662912e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}