reset default eval_batch_size

HFUT-LEC · Feb 25, 2024 · 3f659ef · 3f659ef
1 parent 09c0693
commit 3f659ef
Show file tree

Hide file tree

Showing 6 changed files with 8 additions and 5 deletions.
diff --git a/docs/source/assets/dataflow.jpg b/docs/source/assets/dataflow.jpg
diff --git a/docs/source/features/standard_datamodule.md b/docs/source/features/standard_datamodule.md
@@ -5,7 +5,7 @@ For data module, we provide a standardized design with three protocols (see foll
 - Middle Data Format Protocol
 - Atomic Operation Protocol
 
-![](https://s2.loli.net/2024/02/09/trocOIdR4wxXJDQ.png)
+![](../assets/dataflow.jpg)
 
 The first step of Data Templates is to load the raw data from the hard disk. Then, a series of processing steps are performed to obtain model-friendly data objects. Finally, these data objects are passed on to other modules.
 We simplify the data preparation into three into three stages:

diff --git a/edustudio/datatpl/common/general_datatpl.py b/edustudio/datatpl/common/general_datatpl.py
@@ -541,8 +541,10 @@ def _preprocess_feat(df):
         for col in df.columns:
             col_name, col_type = col.split(":")
             if col_type == 'token':
-                # df[col] = df[col].astype('int64')
-                pass
+                try:
+                    df[col] = df[col].astype('int64')
+                except:
+                    pass
             elif col_type == 'float':
                 df[col] = df[col].astype('float32')
             elif col_type == 'token_seq':

diff --git a/edustudio/traintpl/general_traintpl.py b/edustudio/traintpl/general_traintpl.py
@@ -18,6 +18,7 @@ class GeneralTrainTPL(GDTrainTPL):
         'unsave_best_epoch_pth': True,
         'ignore_metrics_in_train': [],
         'batch_size': 32,
+        'eval_batch_size': 32
     }
 
     def _check_params(self):

diff --git a/examples/single_model/run_cdgk_demo.py b/examples/single_model/run_cdgk_demo.py
@@ -11,7 +11,8 @@
     cfg_file_name=None,
     traintpl_cfg_dict={
         'cls': 'GeneralTrainTPL',
-        'batch_size': 1024
+        'batch_size': 1024,
+        'eval_batch_size': 1024
     },
     datatpl_cfg_dict={
         'cls': 'CDGKDataTPL',

diff --git a/examples/single_model/run_dimkt_demo.py b/examples/single_model/run_dimkt_demo.py
@@ -14,7 +14,6 @@
     },
     traintpl_cfg_dict={
         'cls': 'GeneralTrainTPL',
-        'device': 'cpu',
     },
     modeltpl_cfg_dict={
         'cls': 'DIMKT',