<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<meta name="Generator" content="Microsoft Exchange Server">
<!-- converted from text --><style><!-- .EmailQuote { margin-left: 1pt; padding-left: 4pt; border-left: #800000 2px solid; } --></style>
</head>
<body>
<div>
<div dir="auto">Brilliant, thanks! I have had the time to submit a job, currently waiting for the queue to slow down a bit :)</div>
<div class="x_gmail_extra"><br>
<div class="x_gmail_quote">On 15 May 2020 18:37, Stephan Oepen <oe@ifi.uio.no> wrote:<br type="attribution">
</div>
</div>
</div>
<font size="2"><span style="font-size:10pt;">
<div class="PlainText">hiya,<br>
<br>
> Hi! I’ve been using the nlpl-apex module (dated 2020.01), and I believe it currently uses PyTorch 1.3 — I’ve been trying to train a model using fairseq (<a href="https://github.com/pytorch/fairseq">https://github.com/pytorch/fairseq</a>), and it seems to
require PyTorch 1.4, as well as apex. Do you think it’d be possible to push an upgraded module? Alternatively, a fairseq module would also be cool.<br>
<br>
belatedly, i am closing this thread. the NLPL fairseq module that i<br>
put onto Saga last week also includes an updated version of apex.<br>
<br>
have you had a chance to give that module a shot?<br>
<br>
oe<br>
</div>
</span></font>
</body>
</html>