End of training

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 100.0,
+    "eval_accuracy": 0.991421568627451,
+    "eval_loss": 0.08062744140625,
+    "eval_runtime": 5.6349,
+    "eval_samples_per_second": 1303.311,
+    "eval_steps_per_second": 4.792,
+    "train_loss": 0.4972930908203125,
+    "train_runtime": 820.6819,
+    "train_samples_per_second": 4242.326,
+    "train_steps_per_second": 0.487
+}

config.json

@@ -0,0 +1,49 @@
+{
+  "_name_or_path": "google/vit-base-patch16-224-in21k",
+  "architectures": [
+    "PoptorchPipelinedViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "finetuning_task": "image-classification",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "airplane",
+    "1": "automobile",
+    "2": "bird",
+    "3": "cat",
+    "4": "deer",
+    "5": "dog",
+    "6": "frog",
+    "7": "horse",
+    "8": "ship",
+    "9": "truck"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "airplane": "0",
+    "automobile": "1",
+    "bird": "2",
+    "cat": "3",
+    "deer": "4",
+    "dog": "5",
+    "frog": "6",
+    "horse": "7",
+    "ship": "8",
+    "truck": "9"
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float16",
+  "transformers_version": "4.18.0"
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 100.0,
+    "eval_accuracy": 0.991421568627451,
+    "eval_loss": 0.08062744140625,
+    "eval_runtime": 5.6349,
+    "eval_samples_per_second": 1303.311,
+    "eval_steps_per_second": 4.792
+}

ipu_config.json

@@ -0,0 +1,32 @@
+{
+  "device_iterations": 1,
+  "embedding_serialization_factor": 1,
+  "enable_half_first_order_momentum": true,
+  "enable_half_partials": true,
+  "executable_cache_dir": "./exe_cache",
+  "execute_encoder_on_cpu_for_generation": false,
+  "gradient_accumulation_steps": 128,
+  "inference_device_iterations": 4,
+  "inference_replication_factor": 4,
+  "ipus_per_replica": 4,
+  "layers_per_ipu": [
+    3,
+    3,
+    3,
+    3
+  ],
+  "matmul_proportion": [
+    0.3
+  ],
+  "optimizer_state_offchip": true,
+  "optimum_version": "1.1.1",
+  "output_mode": "final",
+  "profile_dir": "",
+  "recompute_checkpoint_every_layer": true,
+  "replicated_tensor_sharding": true,
+  "replication_factor": 4,
+  "seed": 1337,
+  "sharded_execution_for_inference": false,
+  "transformers_version": "4.18.0",
+  "use_popdist": false
+}

preprocessor_config.json

@@ -0,0 +1,17 @@
+{
+  "do_normalize": true,
+  "do_resize": true,
+  "feature_extractor_type": "ViTFeatureExtractor",
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "size": 224
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c1a5b816c77c323209956e4a7e94d42cc01dda0ceda26f8b00acd999cd88caf
+size 171676081

runs/Jun29_15-30-10_gbnwp-pod015-2.ipu.graphcore.ai/1656513160.6009438/events.out.tfevents.1656513160.gbnwp-pod015-2.ipu.graphcore.ai

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f804a495ec240ebfe2ccadcdc8fbf184fb747f76f815d1bbd09b0c5d7dafb883
+size 3932

runs/Jun29_15-30-10_gbnwp-pod015-2.ipu.graphcore.ai/events.out.tfevents.1656513160.gbnwp-pod015-2.ipu.graphcore.ai

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be928dac95ecd92a556c7adc352f3d7cd55812115d0bfbf3b1a338efd4e3312b
+size 9937

runs/Jun29_15-30-10_gbnwp-pod015-2.ipu.graphcore.ai/events.out.tfevents.1656514106.gbnwp-pod015-2.ipu.graphcore.ai

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89527ca9feddecab0d2c8373e33f972bc8cf0a6fd17be4f301ce0fc085845373
+size 40

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 100.0,
+    "train_loss": 0.4972930908203125,
+    "train_runtime": 820.6819,
+    "train_samples_per_second": 4242.326,
+    "train_steps_per_second": 0.487
+}

trainer_state.json

@@ -0,0 +1,266 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 100.0,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.5,
+      "learning_rate": 1e-05,
+      "loss": 2.2814,
+      "step": 10
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2e-05,
+      "loss": 2.2162,
+      "step": 20
+    },
+    {
+      "epoch": 7.5,
+      "learning_rate": 3e-05,
+      "loss": 2.06,
+      "step": 30
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4e-05,
+      "loss": 1.7632,
+      "step": 40
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 5e-05,
+      "loss": 1.3906,
+      "step": 50
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 6e-05,
+      "loss": 1.0958,
+      "step": 60
+    },
+    {
+      "epoch": 17.5,
+      "learning_rate": 7e-05,
+      "loss": 0.8168,
+      "step": 70
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 8e-05,
+      "loss": 0.6212,
+      "step": 80
+    },
+    {
+      "epoch": 22.5,
+      "learning_rate": 9e-05,
+      "loss": 0.5377,
+      "step": 90
+    },
+    {
+      "epoch": 25.0,
+      "learning_rate": 0.0001,
+      "loss": 0.4409,
+      "step": 100
+    },
+    {
+      "epoch": 27.5,
+      "learning_rate": 9.972609476841367e-05,
+      "loss": 0.4009,
+      "step": 110
+    },
+    {
+      "epoch": 30.0,
+      "learning_rate": 9.890738003669029e-05,
+      "loss": 0.3598,
+      "step": 120
+    },
+    {
+      "epoch": 32.5,
+      "learning_rate": 9.755282581475769e-05,
+      "loss": 0.2784,
+      "step": 130
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 9.567727288213005e-05,
+      "loss": 0.3299,
+      "step": 140
+    },
+    {
+      "epoch": 37.5,
+      "learning_rate": 9.330127018922194e-05,
+      "loss": 0.3236,
+      "step": 150
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 9.045084971874738e-05,
+      "loss": 0.2714,
+      "step": 160
+    },
+    {
+      "epoch": 42.5,
+      "learning_rate": 8.715724127386972e-05,
+      "loss": 0.2654,
+      "step": 170
+    },
+    {
+      "epoch": 45.0,
+      "learning_rate": 8.345653031794292e-05,
+      "loss": 0.254,
+      "step": 180
+    },
+    {
+      "epoch": 47.5,
+      "learning_rate": 7.938926261462366e-05,
+      "loss": 0.2223,
+      "step": 190
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.2671,
+      "step": 200
+    },
+    {
+      "epoch": 52.5,
+      "learning_rate": 7.033683215379002e-05,
+      "loss": 0.2293,
+      "step": 210
+    },
+    {
+      "epoch": 55.0,
+      "learning_rate": 6.545084971874738e-05,
+      "loss": 0.2235,
+      "step": 220
+    },
+    {
+      "epoch": 57.5,
+      "learning_rate": 6.0395584540887963e-05,
+      "loss": 0.2031,
+      "step": 230
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 5.522642316338268e-05,
+      "loss": 0.2034,
+      "step": 240
+    },
+    {
+      "epoch": 62.5,
+      "learning_rate": 5e-05,
+      "loss": 0.189,
+      "step": 250
+    },
+    {
+      "epoch": 65.0,
+      "learning_rate": 4.477357683661734e-05,
+      "loss": 0.1788,
+      "step": 260
+    },
+    {
+      "epoch": 67.5,
+      "learning_rate": 3.960441545911204e-05,
+      "loss": 0.1698,
+      "step": 270
+    },
+    {
+      "epoch": 70.0,
+      "learning_rate": 3.4549150281252636e-05,
+      "loss": 0.2309,
+      "step": 280
+    },
+    {
+      "epoch": 72.5,
+      "learning_rate": 2.9663167846209998e-05,
+      "loss": 0.1637,
+      "step": 290
+    },
+    {
+      "epoch": 75.0,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 0.1669,
+      "step": 300
+    },
+    {
+      "epoch": 77.5,
+      "learning_rate": 2.061073738537635e-05,
+      "loss": 0.1786,
+      "step": 310
+    },
+    {
+      "epoch": 80.0,
+      "learning_rate": 1.6543469682057106e-05,
+      "loss": 0.1792,
+      "step": 320
+    },
+    {
+      "epoch": 82.5,
+      "learning_rate": 1.2842758726130283e-05,
+      "loss": 0.1874,
+      "step": 330
+    },
+    {
+      "epoch": 85.0,
+      "learning_rate": 9.549150281252633e-06,
+      "loss": 0.1805,
+      "step": 340
+    },
+    {
+      "epoch": 87.5,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 0.196,
+      "step": 350
+    },
+    {
+      "epoch": 90.0,
+      "learning_rate": 4.322727117869951e-06,
+      "loss": 0.1615,
+      "step": 360
+    },
+    {
+      "epoch": 92.5,
+      "learning_rate": 2.4471741852423237e-06,
+      "loss": 0.1481,
+      "step": 370
+    },
+    {
+      "epoch": 95.0,
+      "learning_rate": 1.0926199633097157e-06,
+      "loss": 0.1633,
+      "step": 380
+    },
+    {
+      "epoch": 97.5,
+      "learning_rate": 2.7390523158633554e-07,
+      "loss": 0.1673,
+      "step": 390
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 0.0,
+      "loss": 0.1748,
+      "step": 400
+    },
+    {
+      "epoch": 100.0,
+      "step": 400,
+      "total_flos": 1.3814551337974377e+23,
+      "train_loss": 0.4972930908203125,
+      "train_runtime": 820.6819,
+      "train_samples_per_second": 4242.326,
+      "train_steps_per_second": 0.487
+    }
+  ],
+  "max_steps": 400,
+  "num_train_epochs": 100,
+  "start_time": 1656513160.5917685,
+  "total_flos": 1.3814551337974377e+23,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c65d55748ded654a5268d5d3306f7d9bc246dcf963ea6627514b18c0c2311c3
+size 2735