Model save

README.md

@@ -2,8 +2,6 @@
 library_name: transformers
 base_model: andro-flock/b2-classification
 tags:
-- image-segmentation
-- vision
 - generated_from_trainer
 model-index:
 - name: b2-segmentation
@@ -15,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # b2-segmentation
 
-This model is a fine-tuned version of [andro-flock/b2-classification](https://huggingface.co/andro-flock/b2-classification) on the andro-flock/semantic-segment dataset.
+This model is a fine-tuned version of [andro-flock/b2-classification](https://huggingface.co/andro-flock/b2-classification) on an unknown dataset.
 
 ## Model description
 
@@ -35,11 +33,9 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 6e-05
-- train_batch_size: 4
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 1337
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: polynomial
 - training_steps: 500

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 125.0,
-    "total_flos": 4.67800620466176e+17,
-    "train_loss": 0.1090870885848999,
-    "train_runtime": 959.3782,
-    "train_samples_per_second": 4.169,
-    "train_steps_per_second": 0.521
+    "epoch": 500.0,
+    "total_flos": 7.8886713950208e+16,
+    "train_loss": 0.1895984401702881,
+    "train_runtime": 397.1334,
+    "train_samples_per_second": 20.144,
+    "train_steps_per_second": 2.518
 }

config.json

@@ -29,15 +29,17 @@
   ],
   "id2label": {
     "0": "background",
-    "1": "head",
-    "2": "body"
+    "1": "hair",
+    "2": "skin",
+    "3": "cloths"
   },
   "image_size": 224,
   "initializer_range": 0.02,
   "label2id": {
     "background": "0",
-    "body": "2",
-    "head": "1"
+    "cloths": "3",
+    "hair": "1",
+    "skin": "2"
   },
   "layer_norm_eps": 1e-06,
   "mlp_ratios": [

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14842d4a49c211fcec3cd7a79d0a8bcf4f92ee43da4d1e3f6d88c089891de57d
-size 109447092
+oid sha256:3529b865cae264a86f42ccef78802036c949e12444f78c61e7854fdf7b666b73
+size 109450168

runs/Mar07_01-57-13_183eb2dda0db/events.out.tfevents.1741312639.183eb2dda0db.2407.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:874d28aa52d2a3c3cb546be0b17ba5ebab5cc9e5a8a81d0e4aa0ade293332e56
+size 5662

runs/Mar07_02-03-50_183eb2dda0db/events.out.tfevents.1741313034.183eb2dda0db.4118.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:070a33ecc508c8bd126b1c16beba7cb843dc34fc6e031adb61d3f25a7b05b178
+size 6649

runs/Mar07_02-12-31_183eb2dda0db/events.out.tfevents.1741313556.183eb2dda0db.6286.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8760feb8ba6a2009e6e9b958b43b0ee1e7ea7c43830859fd9e075764934533f2
+size 7072

runs/Mar07_02-24-11_183eb2dda0db/events.out.tfevents.1741314258.183eb2dda0db.9202.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b80efeff04198213316d3b79e80557ba0d91de1c96d0bd3a50ddc370dfea0536
+size 6455

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 125.0,
-    "total_flos": 4.67800620466176e+17,
-    "train_loss": 0.1090870885848999,
-    "train_runtime": 959.3782,
-    "train_samples_per_second": 4.169,
-    "train_steps_per_second": 0.521
+    "epoch": 500.0,
+    "total_flos": 7.8886713950208e+16,
+    "train_loss": 0.1895984401702881,
+    "train_runtime": 397.1334,
+    "train_samples_per_second": 20.144,
+    "train_steps_per_second": 2.518
 }

trainer_state.json

@@ -1,97 +1,62 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 125.0,
+  "epoch": 500.0,
   "eval_steps": 500,
-  "global_step": 500,
+  "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 12.5,
-      "grad_norm": 3.6458182334899902,
-      "learning_rate": 5.401e-05,
-      "loss": 0.3174,
-      "step": 50
-    },
-    {
-      "epoch": 25.0,
-      "grad_norm": 2.206636905670166,
+      "epoch": 100.0,
+      "grad_norm": 1.8578704595565796,
       "learning_rate": 4.8020000000000004e-05,
-      "loss": 0.1611,
-      "step": 100
-    },
-    {
-      "epoch": 37.5,
-      "grad_norm": 0.8555470705032349,
-      "learning_rate": 4.203e-05,
-      "loss": 0.1319,
-      "step": 150
-    },
-    {
-      "epoch": 50.0,
-      "grad_norm": 0.6757478713989258,
-      "learning_rate": 3.604e-05,
-      "loss": 0.1038,
+      "loss": 0.4234,
       "step": 200
     },
     {
-      "epoch": 62.5,
-      "grad_norm": 1.159254550933838,
-      "learning_rate": 3.0050000000000002e-05,
-      "loss": 0.0701,
-      "step": 250
+      "epoch": 200.0,
+      "grad_norm": 0.6928507685661316,
+      "learning_rate": 3.604e-05,
+      "loss": 0.1814,
+      "step": 400
     },
     {
-      "epoch": 75.0,
-      "grad_norm": 2.89020037651062,
+      "epoch": 300.0,
+      "grad_norm": 1.1086845397949219,
       "learning_rate": 2.406e-05,
-      "loss": 0.0658,
-      "step": 300
-    },
-    {
-      "epoch": 87.5,
-      "grad_norm": 1.4937611818313599,
-      "learning_rate": 1.807e-05,
-      "loss": 0.0635,
-      "step": 350
+      "loss": 0.1281,
+      "step": 600
     },
     {
-      "epoch": 100.0,
-      "grad_norm": 0.5791894197463989,
+      "epoch": 400.0,
+      "grad_norm": 1.4631288051605225,
       "learning_rate": 1.2079999999999998e-05,
-      "loss": 0.0692,
-      "step": 400
-    },
-    {
-      "epoch": 112.5,
-      "grad_norm": 0.22667507827281952,
-      "learning_rate": 6.0899999999999984e-06,
-      "loss": 0.0585,
-      "step": 450
+      "loss": 0.1144,
+      "step": 800
     },
     {
-      "epoch": 125.0,
-      "grad_norm": 0.2987639307975769,
+      "epoch": 500.0,
+      "grad_norm": 0.9606817364692688,
       "learning_rate": 1e-07,
-      "loss": 0.0497,
-      "step": 500
+      "loss": 0.1007,
+      "step": 1000
     },
     {
-      "epoch": 125.0,
-      "step": 500,
-      "total_flos": 4.67800620466176e+17,
-      "train_loss": 0.1090870885848999,
-      "train_runtime": 959.3782,
-      "train_samples_per_second": 4.169,
-      "train_steps_per_second": 0.521
+      "epoch": 500.0,
+      "step": 1000,
+      "total_flos": 7.8886713950208e+16,
+      "train_loss": 0.1895984401702881,
+      "train_runtime": 397.1334,
+      "train_samples_per_second": 20.144,
+      "train_steps_per_second": 2.518
     }
   ],
-  "logging_steps": 50,
-  "max_steps": 500,
+  "logging_steps": 200,
+  "max_steps": 1000,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 125,
+  "num_train_epochs": 500,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -105,8 +70,8 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.67800620466176e+17,
-  "train_batch_size": 4,
+  "total_flos": 7.8886713950208e+16,
+  "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f63fef78833f2e056701b0761ddf3d0a22cccb53dcde50e0fbfdf6c737ced8db
-size 5496
+oid sha256:a31b223fa03376c8ece09ab0e17c53ee11f53b02c5d17355ccd5503c10c6ee91
+size 5432