End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [NTQAI/pedestrian_gender_recognition](https://huggingface.co/NTQAI/pedestrian_gender_recognition) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1918
-- Accuracy: 0.92
+- Loss: 0.1873
+- Accuracy: 0.93
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 2.88,
+    "eval_accuracy": 0.93,
+    "eval_loss": 0.18732059001922607,
+    "eval_runtime": 68.44,
+    "eval_samples_per_second": 1.461,
+    "eval_steps_per_second": 0.102,
+    "total_flos": 1.784652333412516e+17,
+    "train_loss": 0.3550843662685818,
+    "train_runtime": 4732.18,
+    "train_samples_per_second": 0.507,
+    "train_steps_per_second": 0.008
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.88,
+    "eval_accuracy": 0.93,
+    "eval_loss": 0.18732059001922607,
+    "eval_runtime": 68.44,
+    "eval_samples_per_second": 1.461,
+    "eval_steps_per_second": 0.102
+}

runs/Mar20_01-51-10_e7588fdd1fb5/events.out.tfevents.1710904467.e7588fdd1fb5.860.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a2ba7bfc302d5767968920480b3742c070583f8c8b72c3aad42429311573db5
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.88,
+    "total_flos": 1.784652333412516e+17,
+    "train_loss": 0.3550843662685818,
+    "train_runtime": 4732.18,
+    "train_samples_per_second": 0.507,
+    "train_steps_per_second": 0.008
+}

trainer_state.json

@@ -0,0 +1,78 @@
+{
+  "best_metric": 0.93,
+  "best_model_checkpoint": "mikecho/NTQAI_pedestrian_gender_recognition_v1/checkpoint-25",
+  "epoch": 2.88,
+  "eval_steps": 500,
+  "global_step": 36,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.8,
+      "grad_norm": 5.530843734741211,
+      "learning_rate": 7.3125e-05,
+      "loss": 0.4561,
+      "step": 10
+    },
+    {
+      "epoch": 0.96,
+      "eval_accuracy": 0.92,
+      "eval_loss": 0.2296125739812851,
+      "eval_runtime": 60.9008,
+      "eval_samples_per_second": 1.642,
+      "eval_steps_per_second": 0.115,
+      "step": 12
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 4.8732476234436035,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3486,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.93,
+      "eval_loss": 0.18732059001922607,
+      "eval_runtime": 60.9809,
+      "eval_samples_per_second": 1.64,
+      "eval_steps_per_second": 0.115,
+      "step": 25
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 2.4024288654327393,
+      "learning_rate": 1.6875e-05,
+      "loss": 0.2995,
+      "step": 30
+    },
+    {
+      "epoch": 2.88,
+      "eval_accuracy": 0.92,
+      "eval_loss": 0.19184868037700653,
+      "eval_runtime": 61.0368,
+      "eval_samples_per_second": 1.638,
+      "eval_steps_per_second": 0.115,
+      "step": 36
+    },
+    {
+      "epoch": 2.88,
+      "step": 36,
+      "total_flos": 1.784652333412516e+17,
+      "train_loss": 0.3550843662685818,
+      "train_runtime": 4732.18,
+      "train_samples_per_second": 0.507,
+      "train_steps_per_second": 0.008
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 36,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1.784652333412516e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}