Training in progress, step 3000, checkpoint

checkpoint-3000/config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "openai/clip-vit-large-patch14-336",
+  "architectures": [
+    "CLIPModel"
+  ],
+  "initializer_factor": 1.0,
+  "logit_scale_init_value": 2.6592,
+  "model_type": "clip",
+  "projection_dim": 768,
+  "text_config": {
+    "dropout": 0.0,
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "model_type": "clip_text_model",
+    "num_attention_heads": 12,
+    "projection_dim": 768
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.0.dev0",
+  "vision_config": {
+    "dropout": 0.0,
+    "hidden_size": 1024,
+    "image_size": 336,
+    "intermediate_size": 4096,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768
+  }
+}

checkpoint-3000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42f324b5c5de3a9d4ebbd3345684f58ef75c41d122ad9a00b23755e7d3ca46f8
+size 1711848436

checkpoint-3000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab7399566e6150346704b377e9b5f7c9ae14e347dfcf67ea8e2099a497335f10
+size 3424043887

checkpoint-3000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f9733db299f85727654ada1e2241c2b4f64948c1fd868b50f2037c2a0349da2
+size 14503

checkpoint-3000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a4aba2e3883e8c44a7e42620e0e03aa0200eac0443482811c080e77230b85e2
+size 623

checkpoint-3000/trainer_state.json

@@ -0,0 +1,123 @@
+{
+  "best_metric": 0.7885717749595642,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-3000",
+  "epoch": 0.552689756816507,
+  "eval_steps": 500,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09211495946941783,
+      "grad_norm": 8.437943458557129,
+      "learning_rate": 2.907885040530582e-07,
+      "loss": 0.3017,
+      "step": 500
+    },
+    {
+      "epoch": 0.09211495946941783,
+      "eval_loss": 1.164457082748413,
+      "eval_runtime": 75.2697,
+      "eval_samples_per_second": 16.036,
+      "eval_steps_per_second": 2.006,
+      "step": 500
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "grad_norm": 14.740598678588867,
+      "learning_rate": 2.815770081061164e-07,
+      "loss": 0.2899,
+      "step": 1000
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "eval_loss": 1.01627779006958,
+      "eval_runtime": 76.4934,
+      "eval_samples_per_second": 15.779,
+      "eval_steps_per_second": 1.974,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "grad_norm": 0.17031528055667877,
+      "learning_rate": 2.723655121591746e-07,
+      "loss": 0.273,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "eval_loss": 0.9395213723182678,
+      "eval_runtime": 76.6256,
+      "eval_samples_per_second": 15.752,
+      "eval_steps_per_second": 1.971,
+      "step": 1500
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "grad_norm": 0.0030106802005320787,
+      "learning_rate": 2.6315401621223287e-07,
+      "loss": 0.1818,
+      "step": 2000
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "eval_loss": 0.8942967057228088,
+      "eval_runtime": 76.8089,
+      "eval_samples_per_second": 15.714,
+      "eval_steps_per_second": 1.966,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "grad_norm": 474.4410705566406,
+      "learning_rate": 2.539425202652911e-07,
+      "loss": 0.2094,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "eval_loss": 0.8330363631248474,
+      "eval_runtime": 76.9939,
+      "eval_samples_per_second": 15.677,
+      "eval_steps_per_second": 1.961,
+      "step": 2500
+    },
+    {
+      "epoch": 0.552689756816507,
+      "grad_norm": 314.3533935546875,
+      "learning_rate": 2.447310243183493e-07,
+      "loss": 0.1949,
+      "step": 3000
+    },
+    {
+      "epoch": 0.552689756816507,
+      "eval_loss": 0.7885717749595642,
+      "eval_runtime": 76.7534,
+      "eval_samples_per_second": 15.726,
+      "eval_steps_per_second": 1.967,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 16284,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1079994570482520.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76f7dfc201ccc0039551250d06ef1282a727ca0a51016f874c2eff3c580dbf8e
+size 4847