Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -360,6 +360,7 @@ def create_cluster_dataframes(processed_df):
|
|
360 |
|
361 |
from transformers import GPTNeoForCausalLM, GPT2Tokenizer
|
362 |
def generate_project_proposal(problem_descriptions, location, problem_domain):
|
|
|
363 |
model = GPTNeoForCausalLM.from_pretrained("EleutherAI/gpt-neo-1.3B")
|
364 |
tokenizer = GPT2Tokenizer.from_pretrained("EleutherAI/gpt-neo-1.3B")
|
365 |
|
@@ -379,6 +380,7 @@ def generate_project_proposal(problem_descriptions, location, problem_domain):
|
|
379 |
temperature=0.75)
|
380 |
|
381 |
proposal = tokenizer.decode(output[0], skip_special_tokens=True)
|
|
|
382 |
return proposal
|
383 |
|
384 |
def create_project_proposals(budget_cluster_df, problem_cluster_df, location_clusters, problem_clusters):
|
@@ -467,9 +469,13 @@ def nlp_pipeline(original_df):
|
|
467 |
print("Clustering Done...")
|
468 |
# return processed_df, budget_cluster_df, problem_cluster_df, location_clusters, problem_clusters
|
469 |
|
|
|
|
|
470 |
# # Generate project proposals
|
471 |
location_clusters = dict(enumerate(processed_df['Location_Category_Words'].unique()))
|
472 |
problem_clusters = dict(enumerate(processed_df['Problem_Category_Words'].unique()))
|
|
|
|
|
473 |
project_proposals = create_project_proposals(budget_cluster_df, problem_cluster_df, location_clusters, problem_clusters)
|
474 |
|
475 |
console_messages.append("NLP pipeline completed.")
|
|
|
360 |
|
361 |
from transformers import GPTNeoForCausalLM, GPT2Tokenizer
|
362 |
def generate_project_proposal(problem_descriptions, location, problem_domain):
|
363 |
+
print("Trying to access gpt-neo-1.3B")
|
364 |
model = GPTNeoForCausalLM.from_pretrained("EleutherAI/gpt-neo-1.3B")
|
365 |
tokenizer = GPT2Tokenizer.from_pretrained("EleutherAI/gpt-neo-1.3B")
|
366 |
|
|
|
380 |
temperature=0.75)
|
381 |
|
382 |
proposal = tokenizer.decode(output[0], skip_special_tokens=True)
|
383 |
+
print("Successfully accessed gpt-neo-1.3B and returning")
|
384 |
return proposal
|
385 |
|
386 |
def create_project_proposals(budget_cluster_df, problem_cluster_df, location_clusters, problem_clusters):
|
|
|
469 |
print("Clustering Done...")
|
470 |
# return processed_df, budget_cluster_df, problem_cluster_df, location_clusters, problem_clusters
|
471 |
|
472 |
+
print("location_clusters: ", location_clusters)
|
473 |
+
print("problem_clusters: ", problem_clusters)
|
474 |
# # Generate project proposals
|
475 |
location_clusters = dict(enumerate(processed_df['Location_Category_Words'].unique()))
|
476 |
problem_clusters = dict(enumerate(processed_df['Problem_Category_Words'].unique()))
|
477 |
+
print("location_clusters: ", location_clusters)
|
478 |
+
print("problem_clusters: ", problem_clusters)
|
479 |
project_proposals = create_project_proposals(budget_cluster_df, problem_cluster_df, location_clusters, problem_clusters)
|
480 |
|
481 |
console_messages.append("NLP pipeline completed.")
|