End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: ben81828/CADICA_qwenvl_direction
 tags:
 - llama-factory
+- lora
 - generated_from_trainer
 model-index:
 - name: qwenvl-2B-cadica-direction-then-detect-and-classify-scale6
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # qwenvl-2B-cadica-direction-then-detect-and-classify-scale6
 
-This model is a fine-tuned version of [ben81828/CADICA_qwenvl_direction](https://huggingface.co/ben81828/CADICA_qwenvl_direction) on an unknown dataset.
+This model is a fine-tuned version of [ben81828/CADICA_qwenvl_direction](https://huggingface.co/ben81828/CADICA_qwenvl_direction) on the CADICA狹窄分析選擇題scale6(TRAIN) and the CADICA狹窄分析千問定位但不分類題scale6(TRAIN) datasets.
 It achieves the following results on the evaluation set:
-- Loss: 0.4078
-- Num Input Tokens Seen: 35305984
+- Loss: 0.1728
+- Num Input Tokens Seen: 35316128
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 1.0047267355982274,
+    "eval_loss": 0.17278002202510834,
+    "eval_runtime": 47.6725,
+    "eval_samples_per_second": 1.259,
+    "eval_steps_per_second": 0.315,
+    "num_input_tokens_seen": 35316128,
+    "total_flos": 2329910849044480.0,
+    "train_loss": 0.0003882907692455818,
+    "train_runtime": 103.5396,
+    "train_samples_per_second": 788.104,
+    "train_steps_per_second": 32.838
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0047267355982274,
+    "eval_loss": 0.17278002202510834,
+    "eval_runtime": 47.6725,
+    "eval_samples_per_second": 1.259,
+    "eval_steps_per_second": 0.315,
+    "num_input_tokens_seen": 35316128
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0047267355982274,
+    "num_input_tokens_seen": 35316128,
+    "total_flos": 2329910849044480.0,
+    "train_loss": 0.0003882907692455818,
+    "train_runtime": 103.5396,
+    "train_samples_per_second": 788.104,
+    "train_steps_per_second": 32.838
+}