End of training

README.md

@@ -4,6 +4,8 @@ library_name: peft
 tags:
 - generated_from_trainer
 base_model: HuggingFaceM4/idefics-9b-instruct
+metrics:
+- accuracy
 model-index:
 - name: idefics-9b-instruct-lr-3e-4-city
   results: []
@@ -15,6 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 # idefics-9b-instruct-lr-3e-4-city
 
 This model is a fine-tuned version of [HuggingFaceM4/idefics-9b-instruct](https://huggingface.co/HuggingFaceM4/idefics-9b-instruct) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: nan
+- Accuracy: 0.2713
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.27129337539432175,
+    "eval_loss": NaN,
+    "eval_runtime": 47.8669,
+    "eval_samples": 1,
+    "eval_samples_per_second": 0.021,
+    "eval_steps_per_second": 0.021,
+    "perplexity": NaN,
+    "train_loss": 0.321875,
+    "train_runtime": 84.6245,
+    "train_samples": 1,
+    "train_samples_per_second": 0.059,
+    "train_steps_per_second": 0.059
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.27129337539432175,
+    "eval_loss": NaN,
+    "eval_runtime": 47.8669,
+    "eval_samples": 1,
+    "eval_samples_per_second": 0.021,
+    "eval_steps_per_second": 0.021,
+    "perplexity": NaN
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.321875,
+    "train_runtime": 84.6245,
+    "train_samples": 1,
+    "train_samples_per_second": 0.059,
+    "train_steps_per_second": 0.059
+}

trainer_state.json

@@ -0,0 +1,65 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 10,
+  "global_step": 5,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 1.6094,
+      "step": 1
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": NaN,
+      "learning_rate": 0.0003,
+      "loss": 0.0,
+      "step": 2
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": NaN,
+      "learning_rate": 0.0003,
+      "loss": 0.0,
+      "step": 3
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": NaN,
+      "learning_rate": 0.0003,
+      "loss": 0.0,
+      "step": 4
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": NaN,
+      "learning_rate": 0.0003,
+      "loss": 0.0,
+      "step": 5
+    },
+    {
+      "epoch": 5.0,
+      "step": 5,
+      "total_flos": 2404642160640.0,
+      "train_loss": 0.321875,
+      "train_runtime": 84.6245,
+      "train_samples_per_second": 0.059,
+      "train_steps_per_second": 0.059
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 5,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 2404642160640.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}