Training in progress, step 4500, checkpoint

checkpoint-4500/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-4500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5800e424ea69306e6b1daf1b93cbd56d617be5c6554ecda079de15f449f03847
+size 1711848436

checkpoint-4500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63e6d195933419410dfab6bade151e4aa8086e43d03ce0046ec50a183cd3a2d5
+size 3424043887

checkpoint-4500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53d6607299e954768b3402606a5a5e5f5d35597509d6aebab8b3589af8a48151
+size 14503

checkpoint-4500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ceb6a0722dab923bba32ad2b52778b4bddbeb8bdfe7ba3070c4c3323f6c3abea
+size 623

checkpoint-4500/trainer_state.json

@@ -0,0 +1,168 @@
+{
+  "best_metric": 1.080470085144043,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-4500",
+  "epoch": 0.8290346352247605,
+  "eval_steps": 500,
+  "global_step": 4500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09211495946941783,
+      "grad_norm": 19.808820724487305,
+      "learning_rate": 4.84647506755097e-08,
+      "loss": 0.3758,
+      "step": 500
+    },
+    {
+      "epoch": 0.09211495946941783,
+      "eval_loss": 1.418845295906067,
+      "eval_runtime": 75.2377,
+      "eval_samples_per_second": 16.042,
+      "eval_steps_per_second": 2.007,
+      "step": 500
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "grad_norm": 76.37991333007812,
+      "learning_rate": 4.69295013510194e-08,
+      "loss": 0.4108,
+      "step": 1000
+    },
+    {
+      "epoch": 0.18422991893883567,
+      "eval_loss": 1.3513683080673218,
+      "eval_runtime": 74.9686,
+      "eval_samples_per_second": 16.1,
+      "eval_steps_per_second": 2.014,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "grad_norm": 0.5795987844467163,
+      "learning_rate": 4.539425202652911e-08,
+      "loss": 0.4335,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2763448784082535,
+      "eval_loss": 1.2895005941390991,
+      "eval_runtime": 76.713,
+      "eval_samples_per_second": 15.734,
+      "eval_steps_per_second": 1.968,
+      "step": 1500
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "grad_norm": 0.1463615745306015,
+      "learning_rate": 4.385900270203881e-08,
+      "loss": 0.3436,
+      "step": 2000
+    },
+    {
+      "epoch": 0.36845983787767134,
+      "eval_loss": 1.2414860725402832,
+      "eval_runtime": 76.9749,
+      "eval_samples_per_second": 15.68,
+      "eval_steps_per_second": 1.962,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "grad_norm": 323.8576354980469,
+      "learning_rate": 4.232375337754851e-08,
+      "loss": 0.366,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46057479734708917,
+      "eval_loss": 1.1949334144592285,
+      "eval_runtime": 77.0381,
+      "eval_samples_per_second": 15.668,
+      "eval_steps_per_second": 1.96,
+      "step": 2500
+    },
+    {
+      "epoch": 0.552689756816507,
+      "grad_norm": 339.28533935546875,
+      "learning_rate": 4.078850405305821e-08,
+      "loss": 0.3191,
+      "step": 3000
+    },
+    {
+      "epoch": 0.552689756816507,
+      "eval_loss": 1.1560280323028564,
+      "eval_runtime": 77.0026,
+      "eval_samples_per_second": 15.675,
+      "eval_steps_per_second": 1.961,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6448047162859248,
+      "grad_norm": 60.27859878540039,
+      "learning_rate": 3.925325472856792e-08,
+      "loss": 0.2779,
+      "step": 3500
+    },
+    {
+      "epoch": 0.6448047162859248,
+      "eval_loss": 1.1233855485916138,
+      "eval_runtime": 76.9006,
+      "eval_samples_per_second": 15.696,
+      "eval_steps_per_second": 1.964,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7369196757553427,
+      "grad_norm": 18.268455505371094,
+      "learning_rate": 3.7718005404077616e-08,
+      "loss": 0.2942,
+      "step": 4000
+    },
+    {
+      "epoch": 0.7369196757553427,
+      "eval_loss": 1.0993207693099976,
+      "eval_runtime": 77.2834,
+      "eval_samples_per_second": 15.618,
+      "eval_steps_per_second": 1.954,
+      "step": 4000
+    },
+    {
+      "epoch": 0.8290346352247605,
+      "grad_norm": 505.22723388671875,
+      "learning_rate": 3.6182756079587324e-08,
+      "loss": 0.2615,
+      "step": 4500
+    },
+    {
+      "epoch": 0.8290346352247605,
+      "eval_loss": 1.080470085144043,
+      "eval_runtime": 77.1214,
+      "eval_samples_per_second": 15.651,
+      "eval_steps_per_second": 1.958,
+      "step": 4500
+    }
+  ],
+  "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": 1619632097372520.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-4500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbcece3d46fd39b0ba80eb3e495fdf46c9f891af5b4bddce9de77ad5a467b625
+size 4847