@@ -32,32 +32,29 @@ def run_from_cli(self):
3232 "--dataset-path" , type = str , required = True , help = "Path to the dataset"
3333 )
3434 parser .add_argument (
35- "--dataset-name" , type = str , required = True , help = "Name of the dataset"
36- )
37- parser .add_argument (
38- "--output-path" ,
35+ "--output-dir" ,
3936 type = str ,
4037 required = False ,
41- help = "Path to dump the results" ,
42- default = "output" ,
38+ help = "Directory to dump the results in" ,
4339 )
4440
4541 # Parse the arguments
4642 args = parser .parse_args ()
4743
4844 return self .batch (
4945 dataset_path = args .dataset_path ,
50- dataset_name = args .dataset_name ,
51- output_dir = args .output_path ,
46+ output_dir = args .output_dir ,
5247 )
5348
54- def batch (self , dataset_path : str , dataset_name : str , output_dir : str ):
49+ def batch (self , dataset_path : str , output_dir : str ):
5550 # Load the dataset into a pandas DataFrame
56- df = pd .read_json (dataset_path , orient = "records" )
51+ if dataset_path .endswith (".csv" ):
52+ df = pd .read_csv (dataset_path )
53+ elif dataset_path .endswith (".json" ):
54+ df = pd .read_json (dataset_path , orient = "records" )
5755
5856 # Call the model's run_batch method, passing in the DataFrame
5957 output_df , config = self .run_batch_from_df (df )
60- output_dir = os .path .join (output_dir , dataset_name )
6158 self .write_output_to_directory (output_df , config , output_dir )
6259
6360 def run_batch_from_df (self , df : pd .DataFrame ) -> Tuple [pd .DataFrame , dict ]:
0 commit comments