Download training data from S3

app.py

@@ -10,8 +10,10 @@ import gradio as gr
 import scripts
 import pandas
 
-def greet(name):
-    return "Hello " + name + "!!"
+def download(button):
+    scripts.download_training_data()
+    return gr.Button.update(interactive=True)
+
 
 def train():
     history = scripts.train_models('PlayfulTechnology')
@@ -19,8 +21,10 @@ def train():
 
 
 with gr.Blocks() as trainer:
-    training_button = gr.Button(value="Train models")
+    download_button = gr.Button(value='Doenload training_data')
+    training_button = gr.Button(value="Train models",interactive=False)
     loss_plot = gr.Plot()
+    download_button.click(download,inputs=download_button,outputs=training_button)
     training_button.click(train,inputs=[],outputs=[loss_plot])
     
 trainer.launch()

scripts.py

@@ -19,6 +19,7 @@ import scipy.spatial
 import seaborn
 import tqdm
 import gradio
+import boto3
 
 class SequenceCrossEntropyLoss(torch.nn.Module):
     def __init__(self):
@@ -55,6 +56,16 @@ def clean_question(doc):
         words.append('?')
     return ''.join(words)
 
+def download_training_data():
+    if not os.path.exists('corpora'):
+        os.makedirs('corpora')
+    s3 = boto3.client('s3',
+                      aws_access_key_id=os.environ['AWS_KEY'],
+                      aws_secret_access_key=os.evviron['AWS_SECRET'])
+    for obj in s3.list_objects(Bucket='qarac')['Contents']:
+        filename = obj['Key']
+        s3.download_file('qarac',filename,'corpora/{}'.format(filename))
+
 def prepare_wiki_qa(filename,outfilename):
     data = pandas.read_csv(filename,sep='\t')
     data['QNum']=data['QuestionID'].apply(lambda x: int(x[1:]))