Spaces:
Running
Running
Commit
·
ec7206a
1
Parent(s):
41dd0e4
Update app.py
Browse files
app.py
CHANGED
|
@@ -24,11 +24,19 @@ def extract_abstract(pdf_bytes):
|
|
| 24 |
|
| 25 |
# Function to process text (summarize and convert to speech)
|
| 26 |
def process_text(uploaded_file):
|
| 27 |
-
#
|
| 28 |
-
|
| 29 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
|
| 31 |
-
|
| 32 |
# Function to process text (summarize and convert to speech)
|
| 33 |
#def process_text(uploaded_file):
|
| 34 |
# Extract the file data (byte content) from the uploaded file
|
|
@@ -38,8 +46,6 @@ def process_text(uploaded_file):
|
|
| 38 |
# else:
|
| 39 |
# return "File content could not be retrieved", None
|
| 40 |
|
| 41 |
-
|
| 42 |
-
|
| 43 |
# Generate summary
|
| 44 |
inputs = tokenizer([abstract_text], max_length=1024, return_tensors='pt', truncation=True)
|
| 45 |
summary_ids = model.generate(inputs['input_ids'], num_beams=4, max_length=40, min_length=10, length_penalty=2.0, early_stopping=True, no_repeat_ngram_size=2)
|
|
|
|
| 24 |
|
| 25 |
# Function to process text (summarize and convert to speech)
|
| 26 |
def process_text(uploaded_file):
|
| 27 |
+
# Diagnostic print statements
|
| 28 |
+
print(f"Uploaded file type: {type(uploaded_file)}")
|
| 29 |
+
if isinstance(uploaded_file, dict):
|
| 30 |
+
print("Uploaded file is a dictionary.")
|
| 31 |
+
print(f"Keys available: {uploaded_file.keys()}")
|
| 32 |
+
|
| 33 |
+
# Assuming uploaded_file is a dictionary and contains 'data' key
|
| 34 |
+
try:
|
| 35 |
+
pdf_bytes = uploaded_file["data"]
|
| 36 |
+
except (TypeError, KeyError):
|
| 37 |
+
print("Error accessing 'data' key in uploaded_file")
|
| 38 |
+
return "File content could not be retrieved", None
|
| 39 |
|
|
|
|
| 40 |
# Function to process text (summarize and convert to speech)
|
| 41 |
#def process_text(uploaded_file):
|
| 42 |
# Extract the file data (byte content) from the uploaded file
|
|
|
|
| 46 |
# else:
|
| 47 |
# return "File content could not be retrieved", None
|
| 48 |
|
|
|
|
|
|
|
| 49 |
# Generate summary
|
| 50 |
inputs = tokenizer([abstract_text], max_length=1024, return_tensors='pt', truncation=True)
|
| 51 |
summary_ids = model.generate(inputs['input_ids'], num_beams=4, max_length=40, min_length=10, length_penalty=2.0, early_stopping=True, no_repeat_ngram_size=2)
|