End of training

README.md

@@ -1,8 +1,9 @@
 ---
+base_model: ./clip-roberta
 tags:
 - generated_from_trainer
 datasets:
-- coco_dataset_script
+- ydshieh/coco_dataset_script
 model-index:
 - name: clip-roberta-finetuned
   results: []
@@ -13,7 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # clip-roberta-finetuned
 
-This model was trained from scratch on the coco_dataset_script dataset.
+This model is a fine-tuned version of [./clip-roberta](https://huggingface.co/./clip-roberta) on the ydshieh/coco_dataset_script 2017 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.6986
 
 ## Model description
 

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.6986247301101685,
+    "eval_runtime": 790.7718,
+    "eval_samples_per_second": 31.632,
+    "eval_steps_per_second": 0.494,
+    "train_loss": 0.4136880006540494,
+    "train_runtime": 34968.5642,
+    "train_samples_per_second": 26.142,
+    "train_steps_per_second": 0.409
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "eval_loss": 1.6986247301101685,
+    "eval_runtime": 790.7718,
+    "eval_samples_per_second": 31.632,
+    "eval_steps_per_second": 0.494
+}

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "processor_class": "VisionTextDualEncoderProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 224
+  },
+  "use_square_size": false
+}

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50264": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "processor_class": "VisionTextDualEncoderProcessor",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.4136880006540494,
+    "train_runtime": 34968.5642,
+    "train_samples_per_second": 26.142,
+    "train_steps_per_second": 0.409
+}

trainer_state.json

@@ -0,0 +1,196 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 14286,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8250034999300015e-05,
+      "loss": 1.4008,
+      "step": 500
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.6500069998600034e-05,
+      "loss": 0.9035,
+      "step": 1000
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.475010499790005e-05,
+      "loss": 0.7891,
+      "step": 1500
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.300013999720006e-05,
+      "loss": 0.7111,
+      "step": 2000
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.125017499650007e-05,
+      "loss": 0.6457,
+      "step": 2500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.9500209995800084e-05,
+      "loss": 0.6018,
+      "step": 3000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.77502449951001e-05,
+      "loss": 0.5633,
+      "step": 3500
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.6000279994400116e-05,
+      "loss": 0.5295,
+      "step": 4000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 3.425031499370013e-05,
+      "loss": 0.4894,
+      "step": 4500
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.250034999300014e-05,
+      "loss": 0.4172,
+      "step": 5000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 3.075038499230016e-05,
+      "loss": 0.3773,
+      "step": 5500
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.900041999160017e-05,
+      "loss": 0.3746,
+      "step": 6000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 2.7250454990900182e-05,
+      "loss": 0.3461,
+      "step": 6500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 2.5500489990200198e-05,
+      "loss": 0.3377,
+      "step": 7000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 2.375052498950021e-05,
+      "loss": 0.3257,
+      "step": 7500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 2.2000559988800227e-05,
+      "loss": 0.3138,
+      "step": 8000
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 2.025059498810024e-05,
+      "loss": 0.3026,
+      "step": 8500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.8500629987400252e-05,
+      "loss": 0.289,
+      "step": 9000
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.6750664986700268e-05,
+      "loss": 0.2805,
+      "step": 9500
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 1.500069998600028e-05,
+      "loss": 0.2168,
+      "step": 10000
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 1.3250734985300295e-05,
+      "loss": 0.2078,
+      "step": 10500
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.1500769984600309e-05,
+      "loss": 0.207,
+      "step": 11000
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 9.750804983900321e-06,
+      "loss": 0.1935,
+      "step": 11500
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 8.000839983200337e-06,
+      "loss": 0.1942,
+      "step": 12000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 6.25087498250035e-06,
+      "loss": 0.1903,
+      "step": 12500
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.500909981800364e-06,
+      "loss": 0.1738,
+      "step": 13000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 2.7509449811003783e-06,
+      "loss": 0.1779,
+      "step": 13500
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 1.000979980400392e-06,
+      "loss": 0.1685,
+      "step": 14000
+    },
+    {
+      "epoch": 3.0,
+      "step": 14286,
+      "total_flos": 1.2205375337376e+17,
+      "train_loss": 0.4136880006540494,
+      "train_runtime": 34968.5642,
+      "train_samples_per_second": 26.142,
+      "train_steps_per_second": 0.409
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 14286,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "total_flos": 1.2205375337376e+17,
+  "trial_name": null,
+  "trial_params": null
+}