Update app.py

app.py

@@ -55,6 +55,19 @@ def compute_metrics(eval_pred):
     #Before passing your predictions to compute, you need to convert the predictions to logits (remember all Transformers models return logits):
     return metric.compute(predictions=predictions, references=labels)
 
+#oder mit allen Metriken
+def compute_metrics_alle(eval_pred):
+    metrics = ["accuracy", "recall", "precision", "f1"] #List of metrics to return
+    metric={}
+    for met in metrics:
+       metric[met] = load_metric(met)
+    logits, labels = eval_pred
+    predictions = np.argmax(logits, axis=-1)
+    metric_res={}
+    for met in metrics:
+       metric_res[met]=metric[met].compute(predictions=predictions, references=labels)[met]
+    return metric_res
+
 
 ###################################################################################
 ###################################################################################
@@ -130,7 +143,9 @@ training_args = TrainingArguments(
     overwrite_output_dir = 'True',
     per_device_train_batch_size=batch_size,  #batch_size = 2 for full training
     per_device_eval_batch_size=batch_size,
-    evaluation_strategy = "epoch",  #oder steps
+    evaluation_strategy = "steps",  #oder  
+    logging_strategy="steps",   #oder epoch
+    logging_steps=10,
     learning_rate=2e-5,
     weight_decay=0.01,
     save_total_limit = 2,
@@ -191,6 +206,7 @@ print("done")
 #######################################
 # Load model
 print("load model_neu")
+login(token=os.environ["HF_ACCESS_TOKEN"]) 
 model_name_neu = "alexkueck/test-tis-1"
 #model_neu = trainer.load("test-tis-1")