Model save

README.md

@@ -5,24 +5,24 @@ base_model: vinai/phobert-base
 tags:
 - generated_from_trainer
 model-index:
-- name: clapAI/roberta-base-multilingual-sentiment
+- name: roberta-base-multilingual-sentiment
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# clapAI/roberta-base-multilingual-sentiment
+# roberta-base-multilingual-sentiment
 
 This model is a fine-tuned version of [vinai/phobert-base](https://huggingface.co/vinai/phobert-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.9434
-- Micro F1: 18.2216
-- Micro Precision: 18.2216
-- Micro Recall: 18.2216
-- Macro F1: 9.4327
-- Macro Precision: 17.2520
-- Macro Recall: 13.4610
+- Loss: 1.7158
+- Micro F1: 31.1953
+- Micro Precision: 31.1953
+- Micro Recall: 31.1953
+- Macro F1: 6.7937
+- Macro Precision: 4.4565
+- Macro Recall: 14.2857
 
 ## Model description
 
@@ -51,14 +51,18 @@ The following hyperparameters were used during training:
 - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.01
-- num_epochs: 1.0
+- num_epochs: 5.0
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss | Micro F1 | Micro Precision | Micro Recall | Macro F1 | Macro Precision | Macro Recall |
 |:-------------:|:------:|:----:|:---------------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|
-| 1.9456        | 0.9091 | 5    | 1.9434          | 18.2216  | 18.2216         | 18.2216      | 9.4327   | 17.2520         | 13.4610      |
+| 1.9456        | 1.0    | 6    | 1.9434          | 18.2216  | 18.2216         | 18.2216      | 9.4327   | 17.2520         | 13.4610      |
+| 1.7519        | 2.0    | 12   | 1.9434          | 18.2216  | 18.2216         | 18.2216      | 9.4327   | 17.2520         | 13.4610      |
+| 1.7517        | 3.0    | 18   | 1.9434          | 18.2216  | 18.2216         | 18.2216      | 9.4327   | 17.2520         | 13.4610      |
+| 1.7491        | 4.0    | 24   | 1.7236          | 31.1953  | 31.1953         | 31.1953      | 6.7937   | 4.4565          | 14.2857      |
+| 1.6352        | 4.1818 | 25   | 1.7158          | 31.1953  | 31.1953         | 31.1953      | 6.7937   | 4.4565          | 14.2857      |
 
 
 ### Framework versions

all_results.json

@@ -1,5 +1,5 @@
 {
-    "epoch": 0.9090909090909091,
+    "epoch": 4.181818181818182,
     "eval_f1": 0.09432731429418081,
     "eval_loss": 1.943359375,
     "eval_macro_f1": 9.432731429418082,
@@ -26,8 +26,8 @@
     "test_runtime": 0.3027,
     "test_samples_per_second": 2289.139,
     "test_steps_per_second": 6.606,
-    "train_loss": 1.94560546875,
-    "train_runtime": 12.3942,
-    "train_samples_per_second": 447.627,
-    "train_steps_per_second": 0.403
+    "train_loss": 1.7666796875,
+    "train_runtime": 35.9676,
+    "train_samples_per_second": 771.25,
+    "train_steps_per_second": 0.695
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3f92f4f337ce669ee4c32c92b14424fdb89d85aea6a8ac00f5afb36a7f92301
+oid sha256:9fbc9ff253665d600c1d25222ca96bc4c818f7dc1a2e4c071fd0a7e36e13cb23
 size 270031446

runs/May08_07-42-18_hn-fornix-testing-gpu-platform-2/events.out.tfevents.1746690161.hn-fornix-testing-gpu-platform-2.2694805.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f49db66283a602e2d01baf46b2a1a9ee1b366513145b3ce4660df4898f90449b
-size 7235
+oid sha256:540208cefc1933f6678ef01d47dd8d6430c84a8f4c48e5ac87c465c76f604ccc
+size 9986

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 0.9090909090909091,
-    "train_loss": 1.94560546875,
-    "train_runtime": 12.3942,
-    "train_samples_per_second": 447.627,
-    "train_steps_per_second": 0.403
+    "epoch": 4.181818181818182,
+    "train_loss": 1.7666796875,
+    "train_runtime": 35.9676,
+    "train_samples_per_second": 771.25,
+    "train_steps_per_second": 0.695
 }

trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 5,
-  "best_metric": 18.221574344023324,
-  "best_model_checkpoint": "/data/hungnm/unisentiment/roberta-base-sentiment/checkpoint-5",
-  "epoch": 0.9090909090909091,
+  "best_global_step": 24,
+  "best_metric": 31.195335276967928,
+  "best_model_checkpoint": "/data/hungnm/unisentiment/roberta-base-sentiment/checkpoint-24",
+  "epoch": 4.181818181818182,
   "eval_steps": 500,
-  "global_step": 5,
+  "global_step": 25,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -17,7 +17,7 @@
       "step": 5
     },
     {
-      "epoch": 0.9090909090909091,
+      "epoch": 1.0,
       "eval_loss": 1.943359375,
       "eval_macro_f1": 9.432731429418082,
       "eval_macro_precision": 17.25200652415304,
@@ -25,25 +25,109 @@
       "eval_micro_f1": 18.221574344023324,
       "eval_micro_precision": 18.221574344023324,
       "eval_micro_recall": 18.221574344023324,
-      "eval_runtime": 0.3376,
-      "eval_samples_per_second": 2032.14,
-      "eval_steps_per_second": 5.925,
-      "step": 5
+      "eval_runtime": 0.1703,
+      "eval_samples_per_second": 4027.509,
+      "eval_steps_per_second": 11.742,
+      "step": 6
     },
     {
-      "epoch": 0.9090909090909091,
-      "step": 5,
-      "total_flos": 290224529735680.0,
-      "train_loss": 1.94560546875,
-      "train_runtime": 12.3942,
-      "train_samples_per_second": 447.627,
-      "train_steps_per_second": 0.403
+      "epoch": 1.7272727272727273,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 1.7519,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.943359375,
+      "eval_macro_f1": 9.432731429418082,
+      "eval_macro_precision": 17.25200652415304,
+      "eval_macro_recall": 13.460980331549635,
+      "eval_micro_f1": 18.221574344023324,
+      "eval_micro_precision": 18.221574344023324,
+      "eval_micro_recall": 18.221574344023324,
+      "eval_runtime": 0.1648,
+      "eval_samples_per_second": 4161.491,
+      "eval_steps_per_second": 12.133,
+      "step": 12
+    },
+    {
+      "epoch": 2.5454545454545454,
+      "grad_norm": 0.0,
+      "learning_rate": 0.0,
+      "loss": 1.7517,
+      "step": 15
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.943359375,
+      "eval_macro_f1": 9.432731429418082,
+      "eval_macro_precision": 17.25200652415304,
+      "eval_macro_recall": 13.460980331549635,
+      "eval_micro_f1": 18.221574344023324,
+      "eval_micro_precision": 18.221574344023324,
+      "eval_micro_recall": 18.221574344023324,
+      "eval_runtime": 0.1698,
+      "eval_samples_per_second": 4040.415,
+      "eval_steps_per_second": 11.78,
+      "step": 18
+    },
+    {
+      "epoch": 3.3636363636363638,
+      "grad_norm": 1.8663069009780884,
+      "learning_rate": 5e-05,
+      "loss": 1.7491,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.7236328125,
+      "eval_macro_f1": 6.7936507936507935,
+      "eval_macro_precision": 4.456476468138275,
+      "eval_macro_recall": 14.285714285714285,
+      "eval_micro_f1": 31.195335276967928,
+      "eval_micro_precision": 31.195335276967928,
+      "eval_micro_recall": 31.195335276967928,
+      "eval_runtime": 0.1645,
+      "eval_samples_per_second": 4170.418,
+      "eval_steps_per_second": 12.159,
+      "step": 24
+    },
+    {
+      "epoch": 4.181818181818182,
+      "grad_norm": 2.6474199295043945,
+      "learning_rate": 4.4833833507280884e-05,
+      "loss": 1.6352,
+      "step": 25
+    },
+    {
+      "epoch": 4.181818181818182,
+      "eval_loss": 1.7158203125,
+      "eval_macro_f1": 6.7936507936507935,
+      "eval_macro_precision": 4.456476468138275,
+      "eval_macro_recall": 14.285714285714285,
+      "eval_micro_f1": 31.195335276967928,
+      "eval_micro_precision": 31.195335276967928,
+      "eval_micro_recall": 31.195335276967928,
+      "eval_runtime": 0.1665,
+      "eval_samples_per_second": 4120.855,
+      "eval_steps_per_second": 12.014,
+      "step": 25
+    },
+    {
+      "epoch": 4.181818181818182,
+      "step": 25,
+      "total_flos": 1301189864980480.0,
+      "train_loss": 1.7666796875,
+      "train_runtime": 35.9676,
+      "train_samples_per_second": 771.25,
+      "train_steps_per_second": 0.695
     }
   ],
   "logging_steps": 5,
-  "max_steps": 5,
+  "max_steps": 25,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 5,
   "save_steps": 5.0,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -57,7 +141,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 290224529735680.0,
+  "total_flos": 1301189864980480.0,
   "train_batch_size": 512,
   "trial_name": null,
   "trial_params": null